id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2408.09340 | Xi'an Li | Yilong Hou, Xi'an Li, Jinran Wu and You-Gan Wang | Enhanced BPINN Training Convergence in Solving General and Multi-scale
Elliptic PDEs with Noise | null | null | null | null | cs.LG cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | Bayesian Physics Informed Neural Networks (BPINN) have attracted considerable
attention for inferring the system states and physical parameters of
differential equations according to noisy observations. However, in practice,
Hamiltonian Monte Carlo (HMC) used to estimate the internal parameters of the
solver for BPINN often encounters these troubles including poor performance and
awful convergence for a given step size used to adjust the momentum of those
parameters. To address the convergence of HMC for the BPINN method and extend
its application scope to multi-scale partial differential equations (PDE), we
develop a robust multi-scale BPINN (dubbed MBPINN) method by integrating
multi-scale deep neural networks (MscaleDNN) and the BPINN framework. In this
newly proposed MBPINN method, we reframe HMC with Stochastic Gradient Descent
(SGD) to ensure the most ``likely'' estimation is always provided, and we
configure its solver as a Fourier feature mapping-induced MscaleDNN. This novel
method offers several key advantages: (1) it is more robust than HMC, (2) it
incurs less computational cost than HMC, and (3) it is more flexible for
complex problems. We demonstrate the applicability and performance of the
proposed method through some general Poisson and multi-scale elliptic problems
in one and two-dimensional Euclidean spaces. Our findings indicate that the
proposed method can avoid HMC failures and provide valid results. Additionally,
our method is capable of handling complex elliptic PDE and producing comparable
results for general elliptic PDE under the case of lower signal-to-noise rate.
These findings suggest that our proposed approach has great potential for
physics-informed machine learning for parameter estimation and solution
recovery in the case of ill-posed problems.
| [
{
"version": "v1",
"created": "Sun, 18 Aug 2024 03:20:16 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 20:25:51 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hou",
"Yilong",
""
],
[
"Li",
"Xi'an",
""
],
[
"Wu",
"Jinran",
""
],
[
"Wang",
"You-Gan",
""
]
] |
2408.09780 | Max C. Lemme | Sofia Cruces, Mohit D. Ganeriwala, Jimin Lee, Lukas V\"olkel, Dennis
Braun, Annika Grundmann, Ke Ran, Enrique G. Mar\'in, Holger Kalisch, Michael
Heuken, Andrei Vescan, Joachim Mayer, Andr\'es Godoy, Alwin Daus, Max C.
Lemme | Volatile MoS${_2}$ Memristors with Lateral Silver Ion Migration for
Artificial Neuron Applications | 43 pages | Small Science, 5(5), 2400523, 2025 | 10.1002/smsc.202400523 | null | physics.app-ph cond-mat.mtrl-sci | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Layered two-dimensional (2D) semiconductors have shown enhanced ion migration
capabilities along their van der Waals (vdW) gaps and on their surfaces. This
effect can be employed for resistive switching (RS) in devices for emerging
memories, selectors, and neuromorphic computing. To date, all lateral
molybdenum disulfide (MoS${_2}$)-based volatile RS devices with silver (Ag) ion
migration have been demonstrated using exfoliated, single-crystal MoS${_2}$
flakes requiring a forming step to enable RS. Here, we present volatile RS with
multilayer MoS${_2}$ grown by metal-organic chemical vapor deposition (MOCVD)
with repeatable forming-free operation. The devices show highly reproducible
volatile RS with low operating voltages of approximately 2 V and fast switching
times down to 130 ns considering their micrometer scale dimensions. We
investigate the switching mechanism based on Ag ion surface migration through
transmission electron microscopy, electronic transport modeling, and density
functional theory. Finally, we develop a physics-based compact model and
explore the implementation of our volatile memristors as artificial neurons in
neuromorphic systems.
| [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 08:11:45 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Cruces",
"Sofia",
""
],
[
"Ganeriwala",
"Mohit D.",
""
],
[
"Lee",
"Jimin",
""
],
[
"Völkel",
"Lukas",
""
],
[
"Braun",
"Dennis",
""
],
[
"Grundmann",
"Annika",
""
],
[
"Ran",
"Ke",
""
],
[
"Marín",
"Enrique G.",
""
],
[
"Kalisch",
"Holger",
""
],
[
"Heuken",
"Michael",
""
],
[
"Vescan",
"Andrei",
""
],
[
"Mayer",
"Joachim",
""
],
[
"Godoy",
"Andrés",
""
],
[
"Daus",
"Alwin",
""
],
[
"Lemme",
"Max C.",
""
]
] |
2408.09978 | Bhilahari Jeevanesan | Bhilahari Jeevanesan | Quantum Monte Carlo and Stabilizer States | 10 pages, 3 figures. V2 contains an application to the
Transverse-Field Ising model and an Appendix on Exact Diagonalization | null | null | null | quant-ph cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Quantum-Monte-Carlo technique known as the Stochastic Series Expansion
(SSE) relies on a crucial no-branching condition: the SSE sampling is carried
out in the computational basis, and the no-branching assumption ensures that
superpositions of basis-states do not appear when operators are applied.
Without this proviso, the number of complex amplitudes would grow exponentially
with the number of qubits and would eventually overwhelm the memory and
processing power of a classical computer. However, the action of Clifford group
elements on stabilizer states can be very efficiently described without
resorting to an amplitude description. We explore how stabilizer states allow
an extension of the SSE technique, and we give an example of a toy model that
can be studied in this way. The method is also illustrated using the
Transverse-Field Ising model.
| [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 13:27:15 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 08:56:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jeevanesan",
"Bhilahari",
""
]
] |
2408.10276 | Xiaochen Wang | Xiaochen Wang, Jiaqi Wang, Houping Xiao, Jinghui Chen, Fenglong Ma | FEDKIM: Adaptive Federated Knowledge Injection into Medical Foundation
Models | Accepted by EMNLP'24 Main | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Foundation models have demonstrated remarkable capabilities in handling
diverse modalities and tasks, outperforming conventional artificial
intelligence (AI) approaches that are highly task-specific and
modality-reliant. In the medical domain, however, the development of
comprehensive foundation models is constrained by limited access to diverse
modalities and stringent privacy regulations. To address these constraints,
this study introduces a novel knowledge injection approach, FedKIM, designed to
scale the medical foundation model within a federated learning framework.
FedKIM leverages lightweight local models to extract healthcare knowledge from
private data and integrates this knowledge into a centralized foundation model
using a designed adaptive Multitask Multimodal Mixture Of Experts (M3OE)
module. This method not only preserves privacy but also enhances the model's
ability to handle complex medical tasks involving multiple modalities. Our
extensive experiments across twelve tasks in seven modalities demonstrate the
effectiveness of FedKIM in various settings, highlighting its potential to
scale medical foundation models without direct access to sensitive data.
| [
{
"version": "v1",
"created": "Sat, 17 Aug 2024 15:42:29 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Sep 2024 13:52:51 GMT"
},
{
"version": "v3",
"created": "Tue, 29 Oct 2024 05:18:18 GMT"
},
{
"version": "v4",
"created": "Sun, 4 May 2025 20:14:21 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wang",
"Xiaochen",
""
],
[
"Wang",
"Jiaqi",
""
],
[
"Xiao",
"Houping",
""
],
[
"Chen",
"Jinghui",
""
],
[
"Ma",
"Fenglong",
""
]
] |
2408.10453 | Liu He | Liu He, Yizhi Song, Hejun Huang, Pinxin Liu, Yunlong Tang, Daniel
Aliaga, Xin Zhou | Kubrick: Multimodal Agent Collaborations for Synthetic Video Generation | Accepted by CVPR 2025 AI4CC Workshop | null | null | null | cs.CV cs.GR cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text-to-video generation has been dominated by diffusion-based or
autoregressive models. These novel models provide plausible versatility, but
are criticized for improper physical motion, shading and illumination, camera
motion, and temporal consistency. The film industry relies on manually-edited
Computer-Generated Imagery (CGI) using 3D modeling software. Human-directed 3D
synthetic videos address these shortcomings, but require tight collaboration
between movie makers and 3D rendering experts. We introduce an automatic
synthetic video generation pipeline based on Vision Large Language Model (VLM)
agent collaborations. Given a language description of a video, multiple VLM
agents direct various processes of the generation pipeline. They cooperate to
create Blender scripts which render a video following the given description.
Augmented with Blender-based movie making knowledge, the Director agent
decomposes the text-based video description into sub-processes. For each
sub-process, the Programmer agent produces Python-based Blender scripts based
on function composing and API calling. The Reviewer agent, with knowledge of
video reviewing, character motion coordinates, and intermediate screenshots,
provides feedback to the Programmer agent. The Programmer agent iteratively
improves scripts to yield the best video outcome. Our generated videos show
better quality than commercial video generation models in five metrics on video
quality and instruction-following performance. Our framework outperforms other
approaches in a user study on quality, consistency, and rationality.
| [
{
"version": "v1",
"created": "Mon, 19 Aug 2024 23:31:02 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 07:38:45 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"He",
"Liu",
""
],
[
"Song",
"Yizhi",
""
],
[
"Huang",
"Hejun",
""
],
[
"Liu",
"Pinxin",
""
],
[
"Tang",
"Yunlong",
""
],
[
"Aliaga",
"Daniel",
""
],
[
"Zhou",
"Xin",
""
]
] |
2408.12307 | Fu-Chieh Chang | Yen-Ru Lai, Fu-Chieh Chang, Pei-Yuan Wu | Leveraging Unlabeled Data Sharing through Kernel Function Approximation
in Offline Reinforcement Learning | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Offline reinforcement learning (RL) learns policies from a fixed dataset, but
often requires large amounts of data. The challenge arises when labeled
datasets are expensive, especially when rewards have to be provided by human
labelers for large datasets. In contrast, unlabelled data tends to be less
expensive. This situation highlights the importance of finding effective ways
to use unlabelled data in offline RL, especially when labelled data is limited
or expensive to obtain. In this paper, we present the algorithm to utilize the
unlabeled data in the offline RL method with kernel function approximation and
give the theoretical guarantee. We present various eigenvalue decay conditions
of $\mathcal{H}_k$ which determine the complexity of the algorithm. In summary,
our work provides a promising approach for exploiting the advantages offered by
unlabeled data in offline RL, whilst maintaining theoretical assurances.
| [
{
"version": "v1",
"created": "Thu, 22 Aug 2024 11:31:51 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 07:34:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lai",
"Yen-Ru",
""
],
[
"Chang",
"Fu-Chieh",
""
],
[
"Wu",
"Pei-Yuan",
""
]
] |
2408.13431 | Junjie Liu | Junjie Liu | Face Clustering via Early Stopping and Edge Recall | Insufficient experiments, we hope to withdraw the paper, supplement
and improve it again | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Large-scale face clustering has achieved significant progress, with many
efforts dedicated to learning to cluster large-scale faces with
supervised-learning. However, complex model design and tedious clustering
processes are typical in existing methods. Such limitations result in
infeasible clustering in real-world applications. Reasonable and efficient
model design and training need to be taken into account. Besides, developing
unsupervised face clustering algorithms is crucial, which are more realistic in
real-world applications. In this paper, we propose a novel unsupervised face
clustering algorithm FC-ES and a novel supervised face clustering algorithm
FC-ESER to address these issues. An efficient and effective neighbor-based edge
probability and a novel early stopping strategy are proposed in FC-ES,
guaranteeing the accuracy and recall of large-scale face clustering
simultaneously. Furthermore, to take advantage of supervised learning, a novel
edge recall strategy is proposed in FC-ESER to further recall the edge
connections that are not connected in FC-ES. Extensive experiments on multiple
benchmarks for face, person, and vehicle clustering show that our proposed
FC-ES and FC-ESER significantly outperform previous state-of-the-art methods.
Our code will be available at https://github.com/jumptoliujj/FC-ESER.
| [
{
"version": "v1",
"created": "Sat, 24 Aug 2024 01:53:02 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 12:19:12 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liu",
"Junjie",
""
]
] |
2408.15053 | Alexandru Chirv\u{a}situ L. | Alexandru Chirvasitu, Rafael Dahmen, Karl-Hermann Neeb, Alexander
Schmeding | On the singularities of the exponential function of a semidirect product | 26 pages + references; v4: accepted version, to appear in Classical
and Quantum Gravity | null | null | null | math.GR math-ph math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We show that the Fr\'echet--Lie groups of the form $C^{\infty}(M)\rtimes
\mathbb{R}$ resulting from smooth flows on compact manifolds $M$ fail to be
locally exponential in several cases: when at least one non-periodic orbit is
locally closed, or when the flow restricts to a linear one on an orbit closure
diffeomorphic to a torus. As an application, we prove that the
Bondi--Metzner--Sachs group of symmetries of an asymptotically flat spacetime
is not locally exponential.
| [
{
"version": "v1",
"created": "Tue, 27 Aug 2024 13:29:20 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Oct 2024 07:50:58 GMT"
},
{
"version": "v3",
"created": "Thu, 31 Oct 2024 07:37:21 GMT"
},
{
"version": "v4",
"created": "Sat, 3 May 2025 22:47:24 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chirvasitu",
"Alexandru",
""
],
[
"Dahmen",
"Rafael",
""
],
[
"Neeb",
"Karl-Hermann",
""
],
[
"Schmeding",
"Alexander",
""
]
] |
2408.15186 | Elisa Omodei | J\'ulia Sz\'amely, Alessandro Galeazzi, J\'ulia Koltai and Elisa
Omodei | Easy-access online social media metrics can foster the identification of
misinformation sharing users | null | null | null | null | cs.SI cs.CY physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | Misinformation poses a significant challenge studied extensively by
researchers, yet acquiring data to identify primary sharers is time-consuming
and challenging. To address this, we propose a low-barrier approach to
differentiate social media users who are more likely to share misinformation
from those who are less likely. Leveraging insights from previous studies, we
demonstrate that easy-access online social network metrics -- average daily
tweet count, and account age -- can be leveraged to help identify potential low
factuality content spreaders on X (previously known as Twitter). We find that
higher tweet frequency is positively associated with low factuality in shared
content, while account age is negatively associated with it. We also find that
some of the effects, namely the effect of the number of accounts followed and
the number of tweets produced, differ depending on the number of followers a
user has. Our findings show that relying on these easy-access social network
metrics could serve as a low-barrier approach for initial identification of
users who are more likely to spread misinformation, and therefore contribute to
combating misinformation effectively on social media platforms.
| [
{
"version": "v1",
"created": "Tue, 27 Aug 2024 16:41:13 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 10:03:05 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Számely",
"Júlia",
""
],
[
"Galeazzi",
"Alessandro",
""
],
[
"Koltai",
"Júlia",
""
],
[
"Omodei",
"Elisa",
""
]
] |
2408.15208 | Michael Megrelishvili | Michael Megrelishvili | Lipschitz-free spaces and representations of group actions | 25 pages | null | null | null | math.FA math.DS math.GN | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate actions of topological groups $G$ on Lipschitz-free spaces
$\mathcal{F}(M)$, induced by isometric actions on pointed metric spaces $M$. In
particular, we study the associated dynamical $G$-systems under the weak-star
topology, focusing on the dual action on $\mathrm{Lip}_0(M) = \mathcal{F}(M)^*$
and the bidual $\mathcal{F}(M)^{**}$. Two natural compactification
constructions are considered: the metric compactification of isometric
$G$-spaces over pointed metric spaces, and the Gromov $G$-compactification of
bounded metric $G$-spaces. We show that for every bounded stable metric
$G$-space $(M, d, \mathbf{0})$, the corresponding metric $G$-compactification
defines a weakly almost periodic $G$-flow. We also explore notions of
amenability in this setting, particularly in relation to the metric
compactification, and introduce a topometric version of Lipschitz-free spaces.
| [
{
"version": "v1",
"created": "Tue, 27 Aug 2024 17:16:26 GMT"
},
{
"version": "v2",
"created": "Mon, 3 Mar 2025 17:11:36 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Mar 2025 11:05:51 GMT"
},
{
"version": "v4",
"created": "Thu, 10 Apr 2025 14:25:18 GMT"
},
{
"version": "v5",
"created": "Mon, 5 May 2025 15:58:43 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Megrelishvili",
"Michael",
""
]
] |
2408.15227 | Dan Zhang | C. Goodman, and M. Guzzetti, and C. Hanretty, and L. J. Rosenberg, and
G. Rybka, and J. Sinnis, and D. Zhang, and John Clarke, and I. Siddiqi, and
A. S. Chou, and M. Hollister, and S. Knirck, and A. Sonnenschein, and T. J.
Caligiure, and J. R. Gleason, and A. T. Hipp, and P. Sikivie, and M. E.
Solano, and N. S. Sullivan, and D. B. Tanner, and R. Khatiwada, and G.
Carosi, and C. Cisneros, and N. Du, and N. Robertson, and N. Woollett, and L.
D. Duffy, and C. Boutan, and T. Braine, and E. Lentz, and N. S. Oblath, and
M. S. Taubman, and E. J. Daw, and C. Mostyn, and M. G. Perry, and C. Bartram,
and T. A. Dyson, and S. Ruppert, and M. O. Withers, and C. L. Kuo, and B. T.
McAllister, and J. H. Buckley, and C. Gaikwad, and J. Hoffman, and K. Murch,
and M. Goryachev, and E. Hartman, and A. Quiskamp, and M. E. Tobar | Axion Dark Matter eXperiment around 3.3 {\mu}eV with
Dine-Fischler-Srednicki-Zhitnitsky Discovery Ability | null | Phys. Rev. Lett. 134, 111002 (2025) | 10.1103/PhysRevLett.134.111002 | null | hep-ex | http://creativecommons.org/licenses/by/4.0/ | We report the results of a QCD axion dark matter search with discovery
ability for Dine Fischler Srednicki Zhitnitsky (DFSZ) axions using an axion
haloscope. Sub-Kelvin noise temperatures are reached with an ultra low-noise
Josephson parametric amplifier cooled by a dilution refrigerator. This work
excludes (with a 90% confidence level) DFSZ axions with masses between 3.27 to
3.34 $\mu$eV, assuming a standard halo model with a local energy density of
0.45 GeV/cm$^3$ made up 100% of axions.
| [
{
"version": "v1",
"created": "Tue, 27 Aug 2024 17:46:40 GMT"
},
{
"version": "v2",
"created": "Sun, 10 Nov 2024 20:12:18 GMT"
},
{
"version": "v3",
"created": "Fri, 2 May 2025 21:16:46 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Goodman",
"C.",
""
],
[
"Guzzetti",
"M.",
""
],
[
"Hanretty",
"C.",
""
],
[
"Rosenberg",
"L. J.",
""
],
[
"Rybka",
"G.",
""
],
[
"Sinnis",
"J.",
""
],
[
"Zhang",
"D.",
""
],
[
"Clarke",
"John",
""
],
[
"Siddiqi",
"I.",
""
],
[
"Chou",
"A. S.",
""
],
[
"Hollister",
"M.",
""
],
[
"Knirck",
"S.",
""
],
[
"Sonnenschein",
"A.",
""
],
[
"Caligiure",
"T. J.",
""
],
[
"Gleason",
"J. R.",
""
],
[
"Hipp",
"A. T.",
""
],
[
"Sikivie",
"P.",
""
],
[
"Solano",
"M. E.",
""
],
[
"Sullivan",
"N. S.",
""
],
[
"Tanner",
"D. B.",
""
],
[
"Khatiwada",
"R.",
""
],
[
"Carosi",
"G.",
""
],
[
"Cisneros",
"C.",
""
],
[
"Du",
"N.",
""
],
[
"Robertson",
"N.",
""
],
[
"Woollett",
"N.",
""
],
[
"Duffy",
"L. D.",
""
],
[
"Boutan",
"C.",
""
],
[
"Braine",
"T.",
""
],
[
"Lentz",
"E.",
""
],
[
"Oblath",
"N. S.",
""
],
[
"Taubman",
"M. S.",
""
],
[
"Daw",
"E. J.",
""
],
[
"Mostyn",
"C.",
""
],
[
"Perry",
"M. G.",
""
],
[
"Bartram",
"C.",
""
],
[
"Dyson",
"T. A.",
""
],
[
"Ruppert",
"S.",
""
],
[
"Withers",
"M. O.",
""
],
[
"Kuo",
"C. L.",
""
],
[
"McAllister",
"B. T.",
""
],
[
"Buckley",
"J. H.",
""
],
[
"Gaikwad",
"C.",
""
],
[
"Hoffman",
"J.",
""
],
[
"Murch",
"K.",
""
],
[
"Goryachev",
"M.",
""
],
[
"Hartman",
"E.",
""
],
[
"Quiskamp",
"A.",
""
],
[
"Tobar",
"M. E.",
""
]
] |
2408.15547 | Xiaolei Li | Xiaolei Li, Ryan E. Keeley and Arman Shafieloo | Redshift evolution of the X-ray and UV luminosity relation of quasars:
calibrated results from SNe Ia | 12 pages, 7 figures | The Astrophysical Journal, 2025, 983:141 (9pp) | 10.3847/1538-4357/adc2fe | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quasars could serve as standard candles if the relation between their
ultraviolet and X-ray luminosities can be accurately calibrated. Previously, we
developed a model-independent method to calibrate quasar standard candles using
the distances-redshift relation reconstructed from Type Ia supernova at z<2
using Gaussian process regression. Interestingly, we found that the calibrated
quasar standard candle dataset preferred a deviation from $\Lambda$CDM at
redshifts above z>2. One interpretation of these findings is that the
calibration parameters of the quasar UV-X-ray luminosity relationship evolves
with redshift. In order to test the redshift dependence of the quasar
calibration in a model-independent manner, we divided the quasar sample whose
redshift overlap with the redshift coverage of Pantheon+ Type Ia supernova
compilation into two sub-samples: a low-redshift quasar sub-sample and a
high-redshift quasar sub-sample. Our present results show that there is about a
4$\sigma$ inconsistency between the quasar parameters inferred from the
high-redshift quasar sub-sample and from the low-redshift sub-sample if no
evolution of the quasar relation is considered. This inconsistency suggests the
necessity of considering redshift evolution for the relationship between the
quasars$'$ ultraviolet and X-ray luminosities. We then test an explicit
parametrization of the redshift evolution of the quasar calibration parameters
via $\gamma(z) = \gamma_0+\gamma_1(1+z)$ and $\beta(z)=\beta_0+\beta_1(1+z)$.
Combining this redshift-dependent calibration relationship with the
distance-redshift relationship reconstructed from Pantheon+ supernova
compilation, we find the high-redshift sub-sample and low-redshift sub-sample
become consistent at the 1$\sigma$ level, which means that the parameterized
form of $\gamma(z)$ and $\beta(z)$ works well at describing the evolution of
the quasar calibration parameters.
| [
{
"version": "v1",
"created": "Wed, 28 Aug 2024 05:48:35 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 02:51:24 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Li",
"Xiaolei",
""
],
[
"Keeley",
"Ryan E.",
""
],
[
"Shafieloo",
"Arman",
""
]
] |
2408.16735 | Jan Ebr | Jan Ebr (for the Pierre Auger Collaboration) and Sergey Karpov | FRAM Next Generation at the Pierre Auger Observatory: cloud monitoring
in the age of CMOS cameras | AAM published in Proceedings of AtmoHEAD 2024, CC-BY license | J. Phys.: Conf. Ser. 2985 012009 (2025) | 10.1088/1742-6596/2985/1/012009 | null | astro-ph.IM | http://creativecommons.org/licenses/by/4.0/ | The visibility of stars is often used for cloud detection using all-sky
cameras, which have however only a limited reach and resolution near the
horizon due to the lack of detectable stars. At the Pierre Auger Observatory,
it is also used by the current generation of FRAM robotic telescopes, but --
due to their limited field of view -- only for a small number of selected
showers. Thanks to the recent development in astronomical CMOS cameras, we are
able to propose a new type of device, specifically tailored to the field of
view of the fluorescence detectors (FD) of the Pierre Auger Observatory. The
sub-second readout times available with CMOS cameras allows the efficient use
of short exposures, and so the field of view of one FD can be covered within
half a minute with a resolution and reach sufficient to detect small clouds
with a setup that is significantly smaller, simpler and cheaper than the
current FRAMs. The FRAM Next Generation (framNG) device will be able not only
to detect clouds, but also to assess their optical thickness, provide
information on aerosol extinction, sky brightness and possibly even record
atmospheric phenomena and astrophysical transients. The main challenge lies in
the large data volume produced which necessitates reliable real-time data
processing.
| [
{
"version": "v1",
"created": "Thu, 29 Aug 2024 17:29:55 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 13:08:11 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ebr",
"Jan",
"",
"for the Pierre Auger Collaboration"
],
[
"Karpov",
"Sergey",
""
]
] |
2408.17022 | Philipp Wittenberg | Philipp Ad\"ammer and Philipp Wittenberg and Christian H. Wei{\ss} and
Murat Caner Testik | Nonparametric Monitoring of Spatial Dependence | 38 pages, 9 figures, 19 tables | null | null | null | stat.ME stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In process monitoring, it is common for measurements to be taken regularly or
randomly from different spatial locations in two or three dimensions. While
there are nonparametric methods for process monitoring with such spatial data
to detect changes in the mean, there is a gap in the literature for
nonparametric control charting methods developed to monitor spatial dependence.
This study considers streams of regular, rectangular data sets using spatial
ordinal patterns (SOPs) as a nonparametric method to detect spatial
dependencies. We propose novel SOP control charts, which are distribution-free
and do not require prior Phase-I analysis. To uncover higher-order
dependencies, we develop a new class of statistics that combines SOPs with the
Box-Pierce approach. An extensive simulation study demonstrates the superiority
and effectiveness of our proposed charts over traditional parametric
approaches, particularly when the spatial dependence is nonlinear or bilateral
or when the spatial data contains outliers. The proposed SOP control charts are
illustrated using real-world datasets to detect (i) heavy rainfall in Germany,
(ii) war-related fires in (eastern) Ukraine, and (iii) manufacturing defects in
textile production. This wide range of applications and findings demonstrates
the broad utility of the proposed nonparametric control charts. In addition,
all methods in this study are provided as a publicly available \texttt{Julia}
package on \href{https://github.com/AdaemmerP/OrdinalPatterns.jl}{GitHub} for
further implementations.
| [
{
"version": "v1",
"created": "Fri, 30 Aug 2024 05:29:21 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 11:25:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Adämmer",
"Philipp",
""
],
[
"Wittenberg",
"Philipp",
""
],
[
"Weiß",
"Christian H.",
""
],
[
"Testik",
"Murat Caner",
""
]
] |
2408.17090 | Jingjing Deng PhD | Chen Hu, Hanchi Ren, Jingjing Deng, Xianghua Xie, Xiaoke Ma | FissionVAE: Federated Non-IID Image Generation with Latent Space and
Decoder Decomposition | null | null | null | null | cs.LG cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning is a machine learning paradigm that enables decentralized
clients to collaboratively learn a shared model while keeping all the training
data local. While considerable research has focused on federated image
generation, particularly Generative Adversarial Networks, Variational
Autoencoders have received less attention. In this paper, we address the
challenges of non-IID (independently and identically distributed) data
environments featuring multiple groups of images of different types. Non-IID
data distributions can lead to difficulties in maintaining a consistent latent
space and can also result in local generators with disparate texture features
being blended during aggregation. We thereby introduce FissionVAE that
decouples the latent space and constructs decoder branches tailored to
individual client groups. This method allows for customized learning that
aligns with the unique data distributions of each group. Additionally, we
incorporate hierarchical VAEs and demonstrate the use of heterogeneous decoder
architectures within FissionVAE. We also explore strategies for setting the
latent prior distributions to enhance the decoupling process. To evaluate our
approach, we assemble two composite datasets: the first combines MNIST and
FashionMNIST; the second comprises RGB datasets of cartoon and human faces,
wild animals, marine vessels, and remote sensing images. Our experiments
demonstrate that FissionVAE greatly improves generation quality on these
datasets compared to baseline federated VAE models.
| [
{
"version": "v1",
"created": "Fri, 30 Aug 2024 08:22:30 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 13:51:42 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hu",
"Chen",
""
],
[
"Ren",
"Hanchi",
""
],
[
"Deng",
"Jingjing",
""
],
[
"Xie",
"Xianghua",
""
],
[
"Ma",
"Xiaoke",
""
]
] |
2408.17303 | Andrea Conti | Andrea Conti, Niall T. Macpherson | ${\cal N}=(4,4)$ supersymmetric AdS$_3$ solutions in $d=11$ | 42 pages+ appendices. v3: Minor corrections, new conclusion added | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We derive necessary and sufficient conditions for AdS$_3$ solutions of $d=11$
supergravity to preserve ${\cal N}=(1,1)$ supersymmetry in terms of
G-structures. Such solutions necessarily support an SU(3)-structure on the
internal 8-manifold M$_8$, in terms of which we phrase the conditions for
supersymmetry preservation. We use this to derive the local form of all ${\cal
N}=(4,4)$ supersymmetric AdS$_3$ solutions in $d=11$, for which M$_8$
decomposes as a foliation of a 3-sphere over a 5 dimensional base. There are 3
independent classes, 2 of which preserve the small superconformal algebra and
one preserving its large counterpart for which M$_5$ contains a second
3-sphere. We show that for each solution with large (4,4) supersymmetry there
are two corresponding solutions with small $(4,4)$, one for which M$_5$
maintains its 3-sphere, one where this blows up to $\mathbb{R}^3$ which can be
compactified to $\mathbb{T}^3$. We use our results to construct several new
solutions that lie within our derived classes as well as recovering some
existing solutions.
| [
{
"version": "v1",
"created": "Fri, 30 Aug 2024 13:58:45 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Jan 2025 11:46:08 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 14:25:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Conti",
"Andrea",
""
],
[
"Macpherson",
"Niall T.",
""
]
] |
2409.00067 | Jacob Haqq-Misra | Jacob Haqq-Misra, George Profitiliotis, Ravi Kopparapu | Projections of Earth's technosphere: Scenario modeling, worldbuilding,
and overview of remotely detectable technosignatures | Accepted for publication in Technological Forecasting and Social
Change | null | null | null | physics.soc-ph astro-ph.EP astro-ph.IM physics.pop-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study uses methods from futures studies to develop a set of ten
self-consistent scenarios for Earth's 1000-year future, which can serve as
examples for defining technosignature search strategies. We apply a novel
worldbuilding pipeline that evaluates the dimensions of human needs in each
scenario as a basis for defining the observable properties of the technosphere.
Our scenarios include three with zero-growth stability, two that have collapsed
into a stable state, one that oscillates between growth and collapse, and four
that continue to grow. Only one scenario includes rapid growth that could lead
to interstellar expansion. We examine absorption spectral features for a few
scenarios to illustrate that nitrogen dioxide can serve as a technosignature to
distinguish between present-day Earth, pre-agricultural Earth, and an
industrial 1000-year future Earth. Three of our scenarios are spectrally
indistinguishable from pre-agricultural Earth, even though these scenarios
include expansive technospheres. Up to nine of these scenarios could represent
steady-state examples that could persist for much longer timescales, and it
remains possible that short-duration technospheres could be the most abundant.
Our scenario set provides the basis for further systematic thinking about
technosignature detection as well as for imagining a broad range of
possibilities for Earth's future.
| [
{
"version": "v1",
"created": "Fri, 23 Aug 2024 12:55:19 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 15:08:04 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Haqq-Misra",
"Jacob",
""
],
[
"Profitiliotis",
"George",
""
],
[
"Kopparapu",
"Ravi",
""
]
] |
2409.01833 | Alberto Dom\'inguez Corella | Alberto Dom\'inguez Corella and Tr\'i Minh L\^e | On the growth of nonconvex functionals at strict local minimizers | 19 pages | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | We give new characterizations of growth conditions at strict local
minimizers. The main characterizations are a variant of the so-called tilt
stability property and an analog of the classical Polyak--\L{}ojasiewicz
condition, where the gradient is replaced by linear perturbations. As a
consequence, we derive a tilting principle that relates the stability of
minimizers under linear perturbations to their stability under nonlinear ones.
| [
{
"version": "v1",
"created": "Tue, 3 Sep 2024 12:30:57 GMT"
},
{
"version": "v2",
"created": "Wed, 4 Sep 2024 07:50:40 GMT"
},
{
"version": "v3",
"created": "Mon, 28 Oct 2024 15:38:34 GMT"
},
{
"version": "v4",
"created": "Fri, 2 May 2025 20:07:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Corella",
"Alberto Domínguez",
""
],
[
"Lê",
"Trí Minh",
""
]
] |
2409.01863 | Christine Eagles | Christine Eagles, L\'eo Jimenez | Internality of autonomous algebraic differential equations | 30 pages | null | null | null | math.CA math.LO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This article is interested in internality to the constants of systems of
autonomous algebraic ordinary differential equations. Roughly, this means
determining when can all solutions of such a system be written as a rational
function of finitely many fixed solutions (and their derivatives) and finitely
many constants. If the system is a single order one equation, the answer was
given in an old article of Rosenlicht. In the present work, we completely
answer this question for a large class of systems. As a corollary, we obtain a
necessary condition for the generic solution to be Liouvillian. We then apply
these results to determine exactly when solutions to Poizat equations (a
special case of Li\'enard equations) are internal, answering a question of
Freitag, Jaoui, Marker and Nagloo, and to the classic Lotka-Volterra system,
showing that its generic solutions are almost never Liouvillian.
| [
{
"version": "v1",
"created": "Tue, 3 Sep 2024 13:05:21 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 17:56:29 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Eagles",
"Christine",
""
],
[
"Jimenez",
"Léo",
""
]
] |
2409.02249 | Clarence Protin | Gilda Ferreira and Paulo Oliva and Clarence Lewis Protin | On the Various Translations between Classical, Intuitionistic and Linear
Logic | null | null | null | null | math.LO | http://creativecommons.org/licenses/by/4.0/ | Several different proof translations exist between classical and
intuitionistic logic (negative translations), and intuitionistic and linear
logic (Girard translations). Our aims in this paper are (1) to consider
extensions of intuitionistic linear logic which correspond to each of these
systems, and (2) with this common logical basis, to develop an uniform approach
to devising and simplifying proof translations. As we shall see, through this
process of ``simplification'' we obtain most of the well-known translations in
the literature.
| [
{
"version": "v1",
"created": "Tue, 3 Sep 2024 19:22:06 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 11:10:39 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ferreira",
"Gilda",
""
],
[
"Oliva",
"Paulo",
""
],
[
"Protin",
"Clarence Lewis",
""
]
] |
2409.03423 | Najib Khachiaa | Najib Khachiaa | Discrete Zak Transform and Multi-window Gabor Systems on Discrete
Periodic Sets | null | null | null | null | math.FA | http://creativecommons.org/licenses/by/4.0/ | In this paper, $\mathcal{G}(g,L,M,N)$ denotes a $L-$window Gabor system on a
periodic set $\mathbb{S}$, where $L,M,M\in \mathbb{N}$ and $g=\{g_l\}_{l\in
\mathbb{N}_L}\subset \ell^2(\mathbb{S})$. We characterize which $g$ generates a
complete multi-window Gabor system and a multi-window Gabor frame
$\mathcal{G}(g,L,M,N)$ on $\mathbb{S}$ using the Zak transform. Admissibility
conditions for a periodic set to admit a complete multi--window Gabor system,
multi-window Gabor (Parseval) frame, and multi--window Gabor (orthonormal)
basis $\mathcal{G}(g,L,M,N)$ are given with respect to the parameters $L$, $M$
and $N$.
| [
{
"version": "v1",
"created": "Thu, 5 Sep 2024 11:09:50 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Jan 2025 20:58:22 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 12:30:10 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Khachiaa",
"Najib",
""
]
] |
2409.03638 | Mourad Halla | Mourad Halla | Quantum Natural Gradient with Geodesic Corrections for Small Shallow
Quantum Circuits | null | Physica Scripta, Volume 100, Number 5 (2025), 055121 | 10.1088/1402-4896/add05e | null | quant-ph physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Quantum Natural Gradient (QNG) method enhances optimization in
variational quantum algorithms (VQAs) by incorporating geometric insights from
the quantum state space through the Fubini-Study metric. In this work, we
extend QNG by introducing higher-order integrators and geodesic corrections
using the Riemannian Euler update rule and geodesic equations, deriving an
updated rule for the Quantum Natural Gradient with Geodesic Correction (QNGGC).
We also develop an efficient method for computing the Christoffel symbols
necessary for these corrections, leveraging the parameter-shift rule to enable
direct measurement from quantum circuits. Through theoretical analysis and
practical examples, we demonstrate that QNGGC significantly improves
convergence rates over standard QNG, highlighting the benefits of integrating
geodesic corrections into quantum optimization processes. Our approach paves
the way for more efficient quantum algorithms, leveraging the advantages of
geometric methods.
| [
{
"version": "v1",
"created": "Thu, 5 Sep 2024 15:54:02 GMT"
},
{
"version": "v2",
"created": "Thu, 12 Sep 2024 13:17:05 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Apr 2025 10:30:56 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Halla",
"Mourad",
""
]
] |
2409.03819 | Ashland Knowles | Ashland Knowles, G. Baskaran, R. Ganesh | Quantum loops in the 1T transition metal dichalcogenides | 6 pages, 4 figures, includes 3-page supplement | Phys. Rev. B 111, L020102 (2025) | 10.1103/PhysRevB.111.L020102 | null | cond-mat.str-el | http://creativecommons.org/licenses/by/4.0/ | Loop arrangements and their quantum superpositions describe several
interesting many-particle states. We propose that they also describe bonding in
a class of transition metal dichalcogenides. We present an effective quantum
loop model for monolayers with 1T structure and a d$^2$ valence electron
configuration: materials of the form MX$_2$ (M = Mo, W and X=S, Se, Te) and
AM$'$Y$_2$ (A = Li, Na; M$'$ = V, Nb and Y = O, S, Se). Their t$_{2g}$ orbitals
exhibit strongly directional overlaps between neighbouring atoms, favouring the
formation of valence bonds. A transition metal atom forms two valence bonds,
each with one of its neighbours. When connected, these bonds form loops that
cover the triangular lattice. We construct a minimal Rokhsar-Kivelson-like
model with resonance processes that cut and reconnect loops that run in
proximity. The resulting dynamics is more constrained than in traditional
quantum dimer models, with a `bending' constraint that arises from orbital
structure. In the resulting phase diagram, we find phases that resemble
distorted phases seen in materials, viz., the 1T$'$ and trimerized phases. As a
testable prediction, we propose that a single d$^1$ or d$^3$ impurity will
terminate a loop and give rise to a long-ranged texture. For example, a Ti/Cr
defect in LiVO$_2$ will produce one or more domain walls that propagate outward
from the impurity. We discuss the possibility of a loop liquid phase that can
emerge in these materials.
| [
{
"version": "v1",
"created": "Thu, 5 Sep 2024 18:00:01 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Feb 2025 15:16:27 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Knowles",
"Ashland",
""
],
[
"Baskaran",
"G.",
""
],
[
"Ganesh",
"R.",
""
]
] |
2409.03820 | Robert Moscrop | Philip C. Argyres, Sergio Cecotti, Michele Del Zotto, Mario Martone,
Robert Moscrop, Ben Smith | Allowed Coulomb branch scaling dimensions of four-dimensional
$\mathcal{N} = 2$ SCFTs | 32 pages; v2: 29 pages, major revision, corrected property 5, tables
adjusted accordingly | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A basic datum of a rank-$r$ $\mathcal{N}{=}2$ superconformal field theory
(SCFT) is the $r$-tuple of its Coulomb branch scaling dimensions, i.e., the
scaling dimensions of a set of special protected scalar operators whose vevs
generate the coordinate ring of the Coulomb branch of the theory. It is well
known that when the coordinate ring is freely generated these scaling
dimensions can only take values in a small set of rational numbers. But there
are further constraints on which $r$-tuples of these numbers can appear. The
main aim of this work is to clarify what these are. Along the way we also
compute explicitly the $r$-tuples of allowed scaling dimensions for theories of
ranks $r = 2, 3, 4$.
| [
{
"version": "v1",
"created": "Thu, 5 Sep 2024 18:00:01 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 18:26:51 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Argyres",
"Philip C.",
""
],
[
"Cecotti",
"Sergio",
""
],
[
"Del Zotto",
"Michele",
""
],
[
"Martone",
"Mario",
""
],
[
"Moscrop",
"Robert",
""
],
[
"Smith",
"Ben",
""
]
] |
2409.04372 | Michael Herbst | Michael F. Herbst and Vebj{\o}rn H. Bakkestuen and Andre Laestadius | Kohn-Sham inversion with mathematical guarantees | 6 pages, 4 figures, supporting information see
https://github.com/mfherbst/supporting-my-inversion | null | null | null | physics.chem-ph math-ph math.MP quant-ph | http://creativecommons.org/licenses/by/4.0/ | We use an exact Moreau-Yosida regularized formulation to obtain the
exchange-correlation potential for periodic systems. We reveal a profound
connection between rigorous mathematical principles and efficient numerical
implementation, which marks the first computation of a Moreau-Yosida-based
inversion for physical systems. We develop a mathematically rigorous inversion
algorithm which is demonstrated for representative bulk materials, specifically
bulk silicon, gallium arsenide, and potassium chloride. Our inversion algorithm
allows the construction of rigorous error bounds that we are able to verify
numerically. This unlocks a new pathway to analyze Kohn-Sham inversion methods,
which we expect in turn to foster mathematical approaches for developing
approximate functionals.
| [
{
"version": "v1",
"created": "Fri, 6 Sep 2024 16:08:00 GMT"
},
{
"version": "v2",
"created": "Mon, 3 Mar 2025 20:24:48 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 07:13:40 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Herbst",
"Michael F.",
""
],
[
"Bakkestuen",
"Vebjørn H.",
""
],
[
"Laestadius",
"Andre",
""
]
] |
2409.05314 | Ali Maatouk | Ali Maatouk, Kenny Chirino Ampudia, Rex Ying, Leandros Tassiulas | Tele-LLMs: A Series of Specialized Large Language Models for
Telecommunications | null | null | null | null | cs.IT cs.AI cs.LG math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The emergence of large language models (LLMs) has significantly impacted
various fields, from natural language processing to sectors like medicine and
finance. However, despite their rapid proliferation, the applications of LLMs
in telecommunications remain limited, often relying on general-purpose models
that lack domain-specific specialization. This lack of specialization results
in underperformance, particularly when dealing with telecommunications-specific
technical terminology and their associated mathematical representations. This
paper addresses this gap by first creating and disseminating Tele-Data, a
comprehensive dataset of telecommunications material curated from relevant
sources, and Tele-Eval, a large-scale question-and-answer dataset tailored to
the domain. Through extensive experiments, we explore the most effective
training techniques for adapting LLMs to the telecommunications domain, ranging
from examining the division of expertise across various telecommunications
aspects to employing parameter-efficient techniques. We also investigate how
models of different sizes behave during adaptation and analyze the impact of
their training data on this behavior. Leveraging these findings, we develop and
open-source Tele-LLMs, the first series of language models ranging from 1B to
8B parameters, specifically tailored for telecommunications. Our evaluations
demonstrate that these models outperform their general-purpose counterparts on
Tele-Eval and telecommunications-related literature tasks while retaining their
previously acquired capabilities, thus avoiding the catastrophic forgetting
phenomenon.
| [
{
"version": "v1",
"created": "Mon, 9 Sep 2024 03:58:51 GMT"
},
{
"version": "v2",
"created": "Fri, 13 Sep 2024 23:56:00 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 04:44:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Maatouk",
"Ali",
""
],
[
"Ampudia",
"Kenny Chirino",
""
],
[
"Ying",
"Rex",
""
],
[
"Tassiulas",
"Leandros",
""
]
] |
2409.05929 | Luo Ji | Hongyang Lei, Xiaolong Cheng, Dan Wang, Kun Fan, Qi Qin, Huazhen
Huang, Yetao Wu, Qingqing Gu, Zhonglin Jiang, Yong Chen, Luo Ji | M3-Jepa: Multimodal Alignment via Multi-directional MoE based on the
JEPA framework | 13 pages, 4 figures. Accepted by ICML 2025 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Current multimodal alignment strategies primarily use single or unified
modality encoders, while optimizing the alignment on the original token space.
Such a framework is easy to implement and incorporate with the pretrained
knowledge, but might result in information bias. To deal with such issues, the
joint encoding predictive architecture (JEPA) learns the alignment loss on the
latent space, with a predictor to convert the input encoding to the output
latent space. However, the application of JEPA in multimodal scenarios is
limited so far. In this paper, we introduce M3-Jepa, a scalable multimodal
alignment framework, with the predictor implemented by a multi-directional
mixture of experts (MoE). We demonstrate the framework can maximize the mutual
information with information theory derivations, by alternating the
optimization between different uni-directional tasks. By thoroughly designed
experiments, we show that M3-Jepa can obtain state-of-the-art performance on
different modalities and tasks, generalize to unseen datasets and domains, and
is computationally efficient in training and inference. Our study indicates
that M3-Jepa might provide a new paradigm to self-supervised learning and
open-world modeling.
| [
{
"version": "v1",
"created": "Mon, 9 Sep 2024 10:40:50 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Dec 2024 07:31:18 GMT"
},
{
"version": "v3",
"created": "Fri, 20 Dec 2024 06:37:16 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 16:48:19 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lei",
"Hongyang",
""
],
[
"Cheng",
"Xiaolong",
""
],
[
"Wang",
"Dan",
""
],
[
"Fan",
"Kun",
""
],
[
"Qin",
"Qi",
""
],
[
"Huang",
"Huazhen",
""
],
[
"Wu",
"Yetao",
""
],
[
"Gu",
"Qingqing",
""
],
[
"Jiang",
"Zhonglin",
""
],
[
"Chen",
"Yong",
""
],
[
"Ji",
"Luo",
""
]
] |
2409.05931 | Yuval Wigderson | Yuval Wigderson | Infinitely many minimally non-Ramsey size-linear graphs | 2 pages | null | null | null | math.CO | http://creativecommons.org/licenses/by/4.0/ | A graph $G$ is said to be Ramsey size-linear if $r(G,H) =O_G (e(H))$ for
every graph $H$ with no isolated vertices. Erd\H{o}s, Faudree, Rousseau, and
Schelp observed that $K_4$ is not Ramsey size-linear, but each of its proper
subgraphs is, and they asked whether there exist infinitely many such graphs.
In this short note, we answer this question in the affirmative.
| [
{
"version": "v1",
"created": "Mon, 9 Sep 2024 12:10:38 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:25:39 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wigderson",
"Yuval",
""
]
] |
2409.05975 | Jimeng Shi | Jimeng Shi, Bowen Jin, Jiawei Han, Sundararaman Gopalakrishnan, Giri
Narasimhan | CoDiCast: Conditional Diffusion Model for Global Weather Prediction with
Uncertainty Quantification | 18 pages, 15 figures | null | null | null | cs.LG physics.ao-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate weather forecasting is critical for science and society. Yet,
existing methods have not managed to simultaneously have the properties of high
accuracy, low uncertainty, and high computational efficiency. On one hand, to
quantify the uncertainty in weather predictions, the strategy of ensemble
forecast (i.e., generating a set of diverse predictions) is often employed.
However, traditional ensemble numerical weather prediction (NWP) is
computationally intensive. On the other hand, most existing machine
learning-based weather prediction (MLWP) approaches are efficient and accurate.
Nevertheless, they are deterministic and cannot capture the uncertainty of
weather forecasting. In this work, we propose CoDiCast, a conditional diffusion
model to generate accurate global weather prediction, while achieving
uncertainty quantification with ensemble forecasts and modest computational
cost. The key idea is to simulate a conditional version of the reverse
denoising process in diffusion models, which starts from pure Gaussian noise to
generate realistic weather scenarios for a future time point. Each denoising
step is conditioned on observations from the recent past. Ensemble forecasts
are achieved by repeatedly sampling from stochastic Gaussian noise to represent
uncertainty quantification. CoDiCast is trained on a decade of ERA5 reanalysis
data from the European Centre for Medium-Range Weather Forecasts (ECMWF).
Experimental results demonstrate that our approach outperforms several existing
data-driven methods in accuracy. Our conditional diffusion model, CoDiCast, can
generate 6-day global weather forecasts, at 6-hour steps and $5.625^\circ$
latitude-longitude resolution, for over 5 variables, in about 12 minutes on a
commodity A100 GPU machine with 80GB memory. The open-souced code is provided
at https://github.com/JimengShi/CoDiCast.
| [
{
"version": "v1",
"created": "Mon, 9 Sep 2024 18:18:47 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Sep 2024 20:29:16 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Feb 2025 22:16:55 GMT"
},
{
"version": "v4",
"created": "Fri, 2 May 2025 19:48:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Shi",
"Jimeng",
""
],
[
"Jin",
"Bowen",
""
],
[
"Han",
"Jiawei",
""
],
[
"Gopalakrishnan",
"Sundararaman",
""
],
[
"Narasimhan",
"Giri",
""
]
] |
2409.06541 | Rohan Pramanick | Debajit Bose, Rohan Pramanick, and Tirtha Sankar Ray | Cogenesis of visible and dark matter in a scotogenic model | 12 pages, 7 figures, new constraints included, revised numerical
scans, conclusion unchanged | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | Within a scotogenic neutrino mass model we explore the cogenesis of matter
from the CP violating decay of a heavy $\mathbb{Z}_2$-odd right handed neutrino
that simultaneously populates the visible and a multipartite dark sector. The
quantum of CP violation sets the baryon asymmetry in the visible sector driven
by leptogenesis. The relic density of a sub-GeV scale freeze-in dark matter is
generated by the late time decay of the next to lightest dark particle
dynamically regulated by its interplay with the thermal scattering processes.
| [
{
"version": "v1",
"created": "Tue, 10 Sep 2024 14:15:35 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 14:59:50 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bose",
"Debajit",
""
],
[
"Pramanick",
"Rohan",
""
],
[
"Ray",
"Tirtha Sankar",
""
]
] |
2409.06813 | Leah Weisburn | Leah P. Weisburn, Minsik Cho, Moritz Bensberg, Oinam Romesh Meitei,
Markus Reiher, Troy Van Voorhis | Multiscale Embedding for Quantum Computing | 29 pages, 18 figures, 2 tables; 10 main text figures, 8 supplemental
figures | null | 10.1021/acs.jctc.5c00241 | null | physics.chem-ph cond-mat.str-el quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel multi-scale embedding scheme that links conventional QM/MM
embedding and bootstrap embedding (BE) to allow simulations of large chemical
systems on limited quantum devices. We also propose a mixed-basis BE scheme
that facilitates BE calculations on extended systems using classical computers
with limited memory resources. Benchmark data suggest the combination of these
two strategies as a robust path in attaining the correlation energies of large
realistic systems, combining the proven accuracy of BE with chemical and
biological systems of interest in a lower computational cost method. Due to the
flexible tunability of the resource requirements and systematic fragment
construction, future developments in the realization of quantum computers
naturally offer improved accuracy for multi-scale BE calculations.
| [
{
"version": "v1",
"created": "Tue, 10 Sep 2024 18:41:59 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Mar 2025 18:25:42 GMT"
},
{
"version": "v3",
"created": "Fri, 2 May 2025 21:44:26 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Weisburn",
"Leah P.",
""
],
[
"Cho",
"Minsik",
""
],
[
"Bensberg",
"Moritz",
""
],
[
"Meitei",
"Oinam Romesh",
""
],
[
"Reiher",
"Markus",
""
],
[
"Van Voorhis",
"Troy",
""
]
] |
2409.07219 | Keyu Zhang | Zongxia Liang and Xiang Yu and Keyu Zhang | On time-inconsistent extended mean-field control problems with common
noise | null | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies a class of time-inconsistent mean field control (MFC)
problems in the presence of common noise under non-exponential discount and
joint law dependence of both state and control. We investigate the closed-loop
time-consistent equilibrium strategies for these extended MFC problems and
characterize them through an equilibrium Hamilton-Jacobi-Bellman (HJB) equation
defined on the Wasserstein space. We first apply the results to the
linear-quadratic (LQ) time-inconsistent MFC problems and obtain the existence
of time-consistent equilibria via a comprehensive study of a nonlocal Riccati
system. To illustrate the theoretical findings, two financial applications are
presented. We then examine a class of non-LQ time-inconsistent MFC problems,
for which we contribute the existence of time-consistent equilibria by
analyzing a nonlocal nonlinear partial differential equation.
| [
{
"version": "v1",
"created": "Wed, 11 Sep 2024 12:19:52 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 08:35:39 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liang",
"Zongxia",
""
],
[
"Yu",
"Xiang",
""
],
[
"Zhang",
"Keyu",
""
]
] |
2409.07271 | Weixiang Gao | Weixiang Gao, Yating Zhang and Yifan Xia | CFCPalsy: Facial Image Synthesis with Cross-Fusion Cycle Diffusion Model
for Facial Paralysis Individuals | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Currently, the diagnosis of facial paralysis remains a challenging task,
often relying heavily on the subjective judgment and experience of clinicians,
which can introduce variability and uncertainty in the assessment process. One
promising application in real-life situations is the automatic estimation of
facial paralysis. However, the scarcity of facial paralysis datasets limits the
development of robust machine learning models for automated diagnosis and
therapeutic interventions. To this end, this study aims to synthesize a
high-quality facial paralysis dataset to address this gap, enabling more
accurate and efficient algorithm training. Specifically, a novel Cross-Fusion
Cycle Palsy Expression Generative Model (CFCPalsy) based on the diffusion model
is proposed to combine different features of facial information and enhance the
visual details of facial appearance and texture in facial regions, thus
creating synthetic facial images that accurately represent various degrees and
types of facial paralysis. We have qualitatively and quantitatively evaluated
the proposed method on the commonly used public clinical datasets of facial
paralysis to demonstrate its effectiveness. Experimental results indicate that
the proposed method surpasses state-of-the-art methods, generating more
realistic facial images and maintaining identity consistency.
| [
{
"version": "v1",
"created": "Wed, 11 Sep 2024 13:46:35 GMT"
},
{
"version": "v2",
"created": "Fri, 27 Sep 2024 12:21:26 GMT"
},
{
"version": "v3",
"created": "Thu, 14 Nov 2024 15:36:29 GMT"
},
{
"version": "v4",
"created": "Sat, 3 May 2025 07:07:33 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Gao",
"Weixiang",
""
],
[
"Zhang",
"Yating",
""
],
[
"Xia",
"Yifan",
""
]
] |
2409.07837 | Stanislav \v{Z}ivn\'y | Tamio-Vesa Nakajima and Stanislav \v{Z}ivn\'y | Maximum And- vs. Even-SAT | subsumes arXiv:2402.07863; v2 has more results | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | A (multi)set of literals, called a clause, is strongly satisfied by an
assignment if no literal evaluates to false. Finding an assignment that
maximises the number of strongly satisfied clauses is NP-hard. We present a
simple algorithm that finds, given a set of clauses that admits an assignment
that strongly satisfies a $\rho$-fraction of the clauses, an assignment in
which at least a $\rho$-fraction of the clauses is weakly satisfied, in the
sense that an even number of literals evaluates to false. In particular, this
implies an efficient algorithm for finding an undirected cut of value $\rho$ in
a graph $G$ given that a directed cut of value $\rho$ in $G$ is promised to
exist. A similar argument also gives an efficient algorithm for finding an
acyclic subgraph of $G$ with $\rho$ edges under the same promise.
| [
{
"version": "v1",
"created": "Thu, 12 Sep 2024 08:36:02 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:22:32 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Nakajima",
"Tamio-Vesa",
""
],
[
"Živný",
"Stanislav",
""
]
] |
2409.07916 | Tanima Mondal | T. Mondal, S. Chakraborty, L. Resmi, and D. Bose | Follow-up of Neutron Star Mergers with CTA and Prospects for Joint
Detection with Gravitational-Wave Detectors | 16 pages, 8 figures | Astrophys. J. 983 (2025) 35 | 10.3847/1538-4357/adbd12 | null | astro-ph.HE | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The joint gravitational wave (GW) and electromagnetic observations of the
binary neutron star (BNS) merger GW170817 marked a giant leap in
multi-messenger astrophysics. The extensive observation campaign of the
associated Gamma-Ray Burst (GRB) and its afterglow has strengthened the
hypothesis associating GRBs with BNS mergers and provided insights on mass
ejection, particularly the relativistic outflow launched in BNS mergers. In
this paper, we investigate the joint detection probabilities of BNS mergers by
GW detectors and the upcoming ground-based very-high-energy (VHE) $\gamma$-ray
instrument, the Cherenkov Telescope Array (CTA). Using an empirical relation
that constrains the distance-inclination angle plane, we simulated BNS mergers
detectable in the O5 run of the LIGO/Virgo/Kagra (LVK) network with $300$~Mpc
BNS horizon. Assuming Gaussian structured jets and ignoring large sky
localization challenges of GW detectors, we estimated VHE afterglow detection
probability by CTA. We have explored the afterglow parameter space to identify
conditions favourable for detecting synchrotron self-Compton emission by CTA.
Our study reveals that events viewed at angles $\lesssim3$ times the jet core
angle are detectable by CTA when the initial bulk Lorentz factor at the jet
axis ranges between 100 and 800. We find high kinetic energy ($E_k>10^{50}$
erg), ambient density ($n_0>10^{-1}$ $cm^{-3}$), and energy content in
non-thermal electrons significantly enhance the likelihood of CTA detection
within 300 Mpc. The joint detection rate varies significantly with afterglow
parameter distributions, ranging from $0.003$ to $0.5$ per year.
| [
{
"version": "v1",
"created": "Thu, 12 Sep 2024 10:34:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Mondal",
"T.",
""
],
[
"Chakraborty",
"S.",
""
],
[
"Resmi",
"L.",
""
],
[
"Bose",
"D.",
""
]
] |
2409.08001 | Rotem Assouline | Rotem Assouline | Curvature-Dimension for Autonomous Lagrangians | 97 pages, 4 figures. The previous version of this article was titled
"Magnetic Brunn-Minkowski and Borell-Brascamp-Lieb inequalities on Riemannian
manifolds", and most of its results are contained in the present version as a
special case | null | null | null | math.DG math.MG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a curvature-dimension condition for autonomous Lagrangians on
weighted manifolds, which depends on the Euler-Lagrange dynamics on a single
energy level. By generalizing Klartag's needle decomposition technique to the
Lagrangian setting, we prove that this curvature-dimension condition is
equivalent to displacement convexity of entropy along cost-minimizing
interpolations in an $L^1$ sense, and that it implies various consequences of
lower Ricci curvature bounds, as in the metric setting. As examples we consider
classical and isotropic Lagrangians on Riemannian manifolds. In particular, we
generalize the horocyclic Brunn-Minkowski inequality to complex hyperbolic
space of arbitrary dimension, and present a new Brunn-Minkowski inequality for
contact magnetic geodesics on odd-dimensional spheres.
| [
{
"version": "v1",
"created": "Thu, 12 Sep 2024 12:48:13 GMT"
},
{
"version": "v2",
"created": "Tue, 8 Oct 2024 15:09:41 GMT"
},
{
"version": "v3",
"created": "Sat, 2 Nov 2024 20:40:47 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 06:47:25 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Assouline",
"Rotem",
""
]
] |
2409.08734 | Tobias Wolf | Tobias Wolf, Stefan Kindermann, Elena Resmerita, Luminita Vese | Applications of multiscale hierarchical decomposition to blind
deconvolution | null | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | The blind image deconvolution is a challenging, highly ill-posed nonlinear
inverse problem. We introduce a Multiscale Hierarchical Decomposition Method
(MHDM) that is iteratively solving variational problems with adaptive data and
regularization parameters, towards obtaining finer and finer details of the
unknown kernel and image. We establish convergence of the residual in the
noise-free data case, and then in the noisy data case when the algorithm is
stopped early by means of a discrepancy principle. Fractional Sobolev norms are
employed as regularizers for both kernel and image, with the advantage of
computing the minimizers explicitly in a pointwise manner. In order to break
the notorious symmetry occurring during each minimization step, we enforce a
positivity constraint on the Fourier transform of the kernels. Numerical
comparisons with a single-step variational method and a non-blind MHDM show
that our approach produces comparable results, while less laborious parameter
tuning is necessary at the price of more computations. Additionally, the scale
decomposition of both reconstructed kernel and image provides a meaningful
interpretation of the involved iteration steps.
| [
{
"version": "v1",
"created": "Fri, 13 Sep 2024 11:35:07 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Sep 2024 09:20:57 GMT"
},
{
"version": "v3",
"created": "Fri, 28 Mar 2025 15:45:30 GMT"
},
{
"version": "v4",
"created": "Sun, 4 May 2025 12:35:50 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wolf",
"Tobias",
""
],
[
"Kindermann",
"Stefan",
""
],
[
"Resmerita",
"Elena",
""
],
[
"Vese",
"Luminita",
""
]
] |
2409.09002 | Sudipto Das | Sudipto Das, Sahana Das, and Sudhansu S. Mandal | Enigmatic 2+6/13 Filling Factor: A Prototype Intermittent Topological
State | null | Phys. Rev. B 110, L121303 (2024) | 10.1103/PhysRevB.110.L121303 | null | cond-mat.mes-hall | http://creativecommons.org/licenses/by/4.0/ | Observation of filling factor 6/13 is one of the surprising fractional
quantum Hall states in the second Landau level because, in contrast to the
standard wisdom, the fractions ($\nu < 1/2$) with lower numerators, namely 4
and 5, have not yet been observed. We find that a state indeed forms at
$\nu=6/13$ as an intermittent topological state between two prominent states at
$\nu =1/2$ and $\nu = 2/5$ with lower numerators. Also, we predict that a state
forms at $\nu=5/13$ as an intermittent to $\nu = 2/5$ and $\nu =3/8$. Our
proposed wave functions for $\nu =6/13$ and $5/13$ have excellent overlaps with
the corresponding exact ground state wave functions. The Chern-Simons coupling
matrices deduced from the form of these wave functions are analyzed to predict
the topological properties, which may be experimentally verified.
| [
{
"version": "v1",
"created": "Fri, 13 Sep 2024 17:31:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Das",
"Sudipto",
""
],
[
"Das",
"Sahana",
""
],
[
"Mandal",
"Sudhansu S.",
""
]
] |
2409.09256 | Yifei Xin | Yifei Xin, Zhihong Zhu, Xuxin Cheng, Xusheng Yang, Yuexian Zou | Audio-text Retrieval with Transformer-based Hierarchical Alignment and
Disentangled Cross-modal Representation | Accepted by Interspeech2024 | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most existing audio-text retrieval (ATR) approaches typically rely on a
single-level interaction to associate audio and text, limiting their ability to
align different modalities and leading to suboptimal matches. In this work, we
present a novel ATR framework that leverages two-stream Transformers in
conjunction with a Hierarchical Alignment (THA) module to identify multi-level
correspondences of different Transformer blocks between audio and text.
Moreover, current ATR methods mainly focus on learning a global-level
representation, missing out on intricate details to capture audio occurrences
that correspond to textual semantics. To bridge this gap, we introduce a
Disentangled Cross-modal Representation (DCR) approach that disentangles
high-dimensional features into compact latent factors to grasp fine-grained
audio-text semantic correlations. Additionally, we develop a confidence-aware
(CA) module to estimate the confidence of each latent factor pair and
adaptively aggregate cross-modal latent factors to achieve local semantic
alignment. Experiments show that our THA effectively boosts ATR performance,
with the DCR approach further contributing to consistent performance gains.
| [
{
"version": "v1",
"created": "Sat, 14 Sep 2024 01:54:49 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 14:01:27 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Xin",
"Yifei",
""
],
[
"Zhu",
"Zhihong",
""
],
[
"Cheng",
"Xuxin",
""
],
[
"Yang",
"Xusheng",
""
],
[
"Zou",
"Yuexian",
""
]
] |
2409.09413 | Tadahiro Taniguchi | Tadahiro Taniguchi, Masafumi Oizumi, Noburo Saji, Takato Horii,
Naotsugu Tsuchiya | Constructive Approach to Bidirectional Influence between Qualia
Structure and Language Emergence | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This perspective paper explores the bidirectional influence between language
emergence and the relational structure of subjective experiences, termed qualia
structure, and lays out a constructive approach to the intricate dependency
between the two. We hypothesize that the emergence of languages with
distributional semantics (e.g., syntactic-semantic structures) is linked to the
coordination of internal representations shaped by experience, potentially
facilitating more structured language through reciprocal influence. This
hypothesized mutual dependency connects to recent advancements in AI and symbol
emergence robotics, and is explored within this paper through theoretical
frameworks such as the collective predictive coding. Computational studies show
that neural network-based language models form systematically structured
internal representations, and multimodal language models can share
representations between language and perceptual information. This perspective
suggests that language emergence serves not only as a mechanism creating a
communication tool but also as a mechanism for allowing people to realize
shared understanding of qualitative experiences. The paper discusses the
implications of this bidirectional influence in the context of consciousness
studies, linguistics, and cognitive science, and outlines future constructive
research directions to further explore this dynamic relationship between
language emergence and qualia structure.
| [
{
"version": "v1",
"created": "Sat, 14 Sep 2024 11:03:12 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 03:26:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Taniguchi",
"Tadahiro",
""
],
[
"Oizumi",
"Masafumi",
""
],
[
"Saji",
"Noburo",
""
],
[
"Horii",
"Takato",
""
],
[
"Tsuchiya",
"Naotsugu",
""
]
] |
2409.09724 | Yaning Zhang | Yaning Zhang, Tianyi Wang, Zitong Yu, Zan Gao, Linlin Shen, and
Shengyong Chen | MFCLIP: Multi-modal Fine-grained CLIP for Generalizable Diffusion Face
Forgery Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid development of photo-realistic face generation methods has raised
significant concerns in society and academia, highlighting the urgent need for
robust and generalizable face forgery detection (FFD) techniques. Although
existing approaches mainly capture face forgery patterns using image modality,
other modalities like fine-grained noises and texts are not fully explored,
which limits the generalization capability of the model. In addition, most FFD
methods tend to identify facial images generated by GAN, but struggle to detect
unseen diffusion-synthesized ones. To address the limitations, we aim to
leverage the cutting-edge foundation model, contrastive language-image
pre-training (CLIP), to achieve generalizable diffusion face forgery detection
(DFFD). In this paper, we propose a novel multi-modal fine-grained CLIP
(MFCLIP) model, which mines comprehensive and fine-grained forgery traces
across image-noise modalities via language-guided face forgery representation
learning, to facilitate the advancement of DFFD. Specifically, we devise a
fine-grained language encoder (FLE) that extracts fine global language features
from hierarchical text prompts. We design a multi-modal vision encoder (MVE) to
capture global image forgery embeddings as well as fine-grained noise forgery
patterns extracted from the richest patch, and integrate them to mine general
visual forgery traces. Moreover, we build an innovative plug-and-play sample
pair attention (SPA) method to emphasize relevant negative pairs and suppress
irrelevant ones, allowing cross-modality sample pairs to conduct more flexible
alignment. Extensive experiments and visualizations show that our model
outperforms the state of the arts on different settings like cross-generator,
cross-forgery, and cross-dataset evaluations.
| [
{
"version": "v1",
"created": "Sun, 15 Sep 2024 13:08:59 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 02:34:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhang",
"Yaning",
""
],
[
"Wang",
"Tianyi",
""
],
[
"Yu",
"Zitong",
""
],
[
"Gao",
"Zan",
""
],
[
"Shen",
"Linlin",
""
],
[
"Chen",
"Shengyong",
""
]
] |
2409.09779 | Lijun Zhang | Chengqin Wu and Shuai Yu and Tuyan Luo and Qiuhua Rao and Qingson Hu
and Jingxiang Xu and Lijun Zhang | Underwater Image Enhancement via Dehazing and Color Restoration | null | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Underwater visual imaging is crucial for marine engineering, but it suffers
from low contrast, blurriness, and color degradation, which hinders downstream
analysis. Existing underwater image enhancement methods often treat the haze
and color cast as a unified degradation process, neglecting their inherent
independence while overlooking their synergistic relationship. To overcome this
limitation, we propose a Vision Transformer (ViT)-based network (referred to as
WaterFormer) to improve underwater image quality. WaterFormer contains three
major components: a dehazing block (DehazeFormer Block) to capture the
self-correlated haze features and extract deep-level features, a Color
Restoration Block (CRB) to capture self-correlated color cast features, and a
Channel Fusion Block (CFB) that dynamically integrates these decoupled features
to achieve comprehensive enhancement. To ensure authenticity, a soft
reconstruction layer based on the underwater imaging physics model is included.
Further, a Chromatic Consistency Loss and Sobel Color Loss are designed to
respectively preserve color fidelity and enhance structural details during
network training. Comprehensive experimental results demonstrate that
WaterFormer outperforms other state-of-the-art methods in enhancing underwater
images.
| [
{
"version": "v1",
"created": "Sun, 15 Sep 2024 15:58:20 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 12:21:12 GMT"
},
{
"version": "v3",
"created": "Sun, 4 May 2025 08:27:18 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wu",
"Chengqin",
""
],
[
"Yu",
"Shuai",
""
],
[
"Luo",
"Tuyan",
""
],
[
"Rao",
"Qiuhua",
""
],
[
"Hu",
"Qingson",
""
],
[
"Xu",
"Jingxiang",
""
],
[
"Zhang",
"Lijun",
""
]
] |
2409.10198 | Elena Rufeil Fiori | Elena Rufeil Fiori, Christian Maes and Robbe Vidts | Specific heat of the driven Curie-Weiss model | 9 figures | Journal of Statistical Physics 192, 63 (2025) | null | null | cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Applying a time-periodic magnetic field to the standard ferromagnetic
Curie-Weiss model brings the spin system in a steady out-of-equilibrium
condition. We recall how the hysteresis gets influenced by the amplitude and
the frequency of that field, and how an amplitude- and frequency-dependent
(dynamical) critical temperature can be discerned. The dissipated power
measures the area of the hysteresis loop and changes with temperature. The
excess heat determines a nonequilibrium specific heat giving the quasistatic
response. We compute that specific heat, which appears to diverge at the
critical temperature, quite different from the equilibrium case.
| [
{
"version": "v1",
"created": "Mon, 16 Sep 2024 11:44:59 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Fiori",
"Elena Rufeil",
""
],
[
"Maes",
"Christian",
""
],
[
"Vidts",
"Robbe",
""
]
] |
2409.11464 | Yongda Zhu | Yongda Zhu, Marcia J. Rieke, Zhiyuan Ji, Charlotte Simmonds, Fengwu
Sun, Yang Sun, Stacey Alberts, Rachana Bhatawdekar, Andrew J. Bunker, Phillip
A. Cargile, Stefano Carniani, Anna de Graaff, Kevin Hainline, Jakob M.
Helton, Gareth C. Jones, Jianwei Lyu, George H. Rieke, Pierluigi Rinaldi,
Brant Robertson, Jan Scholtz, Hannah \"Ubler, Christina C. Williams,
Christopher N. A. Willmer | A Systematic Search for Galaxies with Extended Emission Line and
Potential Outflows in JADES Medium-Band Images | 23 pages, 12 figures, 2 tables. Accepted for publication in ApJ | null | null | null | astro-ph.GA astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | For the first time, we present a systematic search for galaxies with extended
emission line and potential outflow features using \textit{JWST} medium-band
images in the GOODS-S field. This is done by comparing the morphology in
medium-band images to adjacent continuum and UV bands. We look for galaxies
that have a maximum extent 50\% larger, an excess area 30\% greater, or an axis
ratio difference of more than 0.3 in the medium band compared to the reference
bands. After visual inspection, we find 326 candidate galaxies at $1.4 < z <
8.4$, with a peak in the population near cosmic noon, benefiting from the good
coverage of the medium-band filters. By fitting their SEDs, we find that the
candidate galaxies are at least 20\% more bursty in their star-forming activity
and have 50\% more young stellar populations compared to a control sample
selected based on the continuum band flux. Additionally, these candidates
exhibit a significantly higher production rate of ionizing photons. We further
find that candidates hosting known AGN produce extended emission that is more
anisotropic compared to non-AGN candidates. A few of our candidates have been
spectroscopically confirmed to have prominent outflow signatures through
NIRSpec observations, showcasing the robustness of the photometric selection.
Future spectroscopic follow-up will better help verify and characterize the
kinematics and chemical properties of these systems.
| [
{
"version": "v1",
"created": "Tue, 17 Sep 2024 18:00:02 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Feb 2025 01:04:36 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 16:53:01 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhu",
"Yongda",
""
],
[
"Rieke",
"Marcia J.",
""
],
[
"Ji",
"Zhiyuan",
""
],
[
"Simmonds",
"Charlotte",
""
],
[
"Sun",
"Fengwu",
""
],
[
"Sun",
"Yang",
""
],
[
"Alberts",
"Stacey",
""
],
[
"Bhatawdekar",
"Rachana",
""
],
[
"Bunker",
"Andrew J.",
""
],
[
"Cargile",
"Phillip A.",
""
],
[
"Carniani",
"Stefano",
""
],
[
"de Graaff",
"Anna",
""
],
[
"Hainline",
"Kevin",
""
],
[
"Helton",
"Jakob M.",
""
],
[
"Jones",
"Gareth C.",
""
],
[
"Lyu",
"Jianwei",
""
],
[
"Rieke",
"George H.",
""
],
[
"Rinaldi",
"Pierluigi",
""
],
[
"Robertson",
"Brant",
""
],
[
"Scholtz",
"Jan",
""
],
[
"Übler",
"Hannah",
""
],
[
"Williams",
"Christina C.",
""
],
[
"Willmer",
"Christopher N. A.",
""
]
] |
2409.11525 | Justin Philip Tuazon | Justin Philip Tuazon, Gia Mizrane Abubo, Joemari Olea | Pairwise Target Rotation for Factor Models | null | null | null | null | stat.ME | http://creativecommons.org/licenses/by/4.0/ | Factor analysis is a way to characterize the relationships between many
manifest variables in terms of a smaller number of latent variables (i.e.,
factors). Particularly, in exploratory factor analysis (EFA), researchers
consider various factor models by exploiting rotational indeterminacy to
uncover underlying structures and identify factors. Generally, the success of
EFA lies with the factor models' interpretabilities, which can be difficult to
achieve or measure. To help address this problem, a new interpretability
criterion is constructed, as well as a rotation method based on it that is
called pairwise target rotation or priorimax. Pairwise target rotation allows
for an intuitive yet flexible way of incorporating arbitrary prior information,
such as semantics, in factor rotations, which can help the researcher perform
EFA more effectively. The implementation of the proposed method is written in
Python 3 and is made available together with several helper functions through
the package interpretablefa on the Python Package Index. A demonstration of the
method is provided using data on Experiences in Close Relationships Scale
obtained from the Open-Source Psychometrics Project.
| [
{
"version": "v1",
"created": "Tue, 17 Sep 2024 19:50:01 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Sep 2024 03:23:14 GMT"
},
{
"version": "v3",
"created": "Sun, 22 Sep 2024 14:01:09 GMT"
},
{
"version": "v4",
"created": "Mon, 2 Dec 2024 13:01:46 GMT"
},
{
"version": "v5",
"created": "Wed, 4 Dec 2024 02:28:32 GMT"
},
{
"version": "v6",
"created": "Sun, 29 Dec 2024 03:39:05 GMT"
},
{
"version": "v7",
"created": "Sun, 4 May 2025 18:30:51 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Tuazon",
"Justin Philip",
""
],
[
"Abubo",
"Gia Mizrane",
""
],
[
"Olea",
"Joemari",
""
]
] |
2409.11529 | Lukas Schynol | Lukas Schynol, Marius Pesavento | Adaptive Anomaly Detection in Network Flows with Low-Rank Tensor
Decompositions and Deep Unrolling | 18 pages, 7 figures | in IEEE Open Journal of Signal Processing, vol. 6, pp. 417-433,
2025 | 10.1109/OJSP.2025.3549350 | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anomaly detection (AD) is increasingly recognized as a key component for
ensuring the resilience of future communication systems. While deep learning
has shown state-of-the-art AD performance, its application in critical systems
is hindered by concerns regarding training data efficiency, domain adaptation
and interpretability. This work considers AD in network flows using incomplete
measurements, leveraging a robust tensor decomposition approach and deep
unrolling techniques to address these challenges. We first propose a novel
block-successive convex approximation algorithm based on a regularized
model-fitting objective where the normal flows are modeled as low-rank tensors
and anomalies as sparse. An augmentation of the objective is introduced to
decrease the computational cost. We apply deep unrolling to derive a novel deep
network architecture based on our proposed algorithm, treating the
regularization parameters as learnable weights. Inspired by Bayesian
approaches, we extend the model architecture to perform online adaptation to
per-flow and per-time-step statistics, improving AD performance while
maintaining a low parameter count and preserving the problem's permutation
equivariances. To optimize the deep network weights for detection performance,
we employ a homotopy optimization approach based on an efficient approximation
of the area under the receiver operating characteristic curve. Extensive
experiments on synthetic and real-world data demonstrate that our proposed deep
network architecture exhibits a high training data efficiency, outperforms
reference methods, and adapts seamlessly to varying network topologies.
| [
{
"version": "v1",
"created": "Tue, 17 Sep 2024 19:59:57 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 17:23:02 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Schynol",
"Lukas",
""
],
[
"Pesavento",
"Marius",
""
]
] |
2409.11674 | Shaojie Hu | Xuhui Su, Dawei Wang, Shaojie Hu | Distinguishing Backward Volume Magnetostatic Spin Wave Vectors via the
Spin Wave Doppler Effect | 14 pages, 6 figures | null | null | null | physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spin waves (SWs) and their quanta, magnons, are essential to achieving
low-power information transmission in future spintronic devices. Backward
volume magnetostatic spin waves (BVMSWs) exhibit a unique dispersion
relationship: one frequency corresponding to two distinct wave vectors. At low
wave numbers, dipole-dipole interactions dominate, resulting in negative group
velocities, whereas at high wave numbers, exchange interactions prevail,
producing positive group velocities. This dual behavior complicates wave vector
identification and obscures intrinsic spin-wave interactions. In this study, we
propose an approach based on the spin wave Doppler effect to effectively
distinguish different wave vectors. At low wave numbers, the inverse Doppler
effect occurs due to antiparallel phase and group velocities, while at high
wave numbers, a normal Doppler effect emerges from parallel velocities. This
method not only clarifies the underlying spin-wave interactions but also help
mitigate serious interference issues in the design of spin logic circuits.
| [
{
"version": "v1",
"created": "Wed, 18 Sep 2024 03:28:18 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 07:49:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Su",
"Xuhui",
""
],
[
"Wang",
"Dawei",
""
],
[
"Hu",
"Shaojie",
""
]
] |
2409.11800 | Matthias Hofmann | Matthias Hofmann, Matthias T\"aufer | Graph structure of the nodal set and bounds on the number of critical
points of eigenfunctions on Riemannian manifolds | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this article, we illustrate and draw connections between the geometry of
zero sets of eigenfunctions, graph theory and the vanishing order of
eigenfunctions. We identify the nodal set of an eigenfunction of the Laplacian
(with smooth potential) on a compact, two-dmensional Riemannian manifolds, that
is on Riemannian surfaces, as an embedded metric graph and then use tools from
elementary graph theory in order to estimate the number of critical points in
the nodal set of the $k$-th eigenfunction and the sum of vanishing orders at
critical points in terms of $k$ and the Euler-Poincar\'e characteristic of the
surface.
| [
{
"version": "v1",
"created": "Wed, 18 Sep 2024 08:31:30 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 10:04:24 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hofmann",
"Matthias",
""
],
[
"Täufer",
"Matthias",
""
]
] |
2409.12232 | Helena Treiber | Helena Treiber, Jenny Greene, John R. Weaver, Tim B. Miller, Lukas J.
Furtak, David J. Setton, Bingjie Wang, Anna de Graaff, Rachel Bezanson,
Gabriel Brammer, Sam E. Cutler, Pratika Dayal, Robert Feldmann, Seiji
Fujimoto, Andy D. Goulding, Vasily Kokorev, Ivo Labbe, Joel Leja, Danilo
Marchesini, Themiya Nanayakkara, Erica Nelson, Richard Pan, Sedona H. Price,
Jared Siegel, Katherine Suess, Katherine Whitaker | UNCOVERing the High-redshift AGN Population among Extreme UV Line
Emitters | Accepted to ApJ | null | 10.3847/1538-4357/adc38f | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | JWST has revealed diverse new populations of high-redshift ($z\sim4-11$) AGN
and extreme star-forming galaxies that challenge current models. In this paper,
we use rest-frame UV emission-line diagnostics to identify AGN candidates and
other exceptional ionizing sources, complementing previous studies
predominantly focused on broad-line AGN. From a parent sample of 205
$\mathrm{z_{spec}}>3$ UNCOVER galaxies with NIRSpec/PRISM follow-up, we
identify 12 C IV, He II, and C III] emitters. Three of these galaxies also
exhibit clear N III] and/or N IV] lines. Leveraging the combined rest-optical
and UV coverage of PRISM, we limit the emission-line model space using the
sample's [O III]/H$\beta$ distribution, significantly decreasing the overlap
between AGN and star-formation models in the UV diagnostics. We then find that
the five He II emitters are the strongest AGN candidates, with further support
from two [Ne V] detections and one X-ray detection from Chandra. Our Balmer
line fits also reveal one new broad-line AGN at z=6.87. We cannot robustly
quantify the AGN fraction in this sample, but we note that close to 20% of
$\mathrm{M_{*}>2\times10^{9}\,M_{\odot}}$ parent sample galaxies are AGN
candidates. The lower-mass line emitters, which are consistent with both AGN
and star-forming photoionization models, have more compact sizes and higher
specific star formation rates than the parent sample. Higher-resolution and
deeper data on these UV line emitters should provide much stronger constraints
on the obscured AGN fraction at $z > 3$.
| [
{
"version": "v1",
"created": "Wed, 18 Sep 2024 18:00:02 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 22:44:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Treiber",
"Helena",
""
],
[
"Greene",
"Jenny",
""
],
[
"Weaver",
"John R.",
""
],
[
"Miller",
"Tim B.",
""
],
[
"Furtak",
"Lukas J.",
""
],
[
"Setton",
"David J.",
""
],
[
"Wang",
"Bingjie",
""
],
[
"de Graaff",
"Anna",
""
],
[
"Bezanson",
"Rachel",
""
],
[
"Brammer",
"Gabriel",
""
],
[
"Cutler",
"Sam E.",
""
],
[
"Dayal",
"Pratika",
""
],
[
"Feldmann",
"Robert",
""
],
[
"Fujimoto",
"Seiji",
""
],
[
"Goulding",
"Andy D.",
""
],
[
"Kokorev",
"Vasily",
""
],
[
"Labbe",
"Ivo",
""
],
[
"Leja",
"Joel",
""
],
[
"Marchesini",
"Danilo",
""
],
[
"Nanayakkara",
"Themiya",
""
],
[
"Nelson",
"Erica",
""
],
[
"Pan",
"Richard",
""
],
[
"Price",
"Sedona H.",
""
],
[
"Siegel",
"Jared",
""
],
[
"Suess",
"Katherine",
""
],
[
"Whitaker",
"Katherine",
""
]
] |
2409.12895 | Diana Vaman | Yuchen Du, Siddarth Ajith, Ravisankar Rajagopal and Diana Vaman | Worldline Proof of Eikonal Exponentiation | 40 pages, references added | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, working in the eikonal approximation, we present a proof for
the exponentiation of the 2-body eikonal phase to {\it all orders in the
eikonal expansion}, for scalar particles interacting electromagnetically or
gravitationally. The proof is based on the worldline formalism, which is an
alternative, first quantized method to the standard QFT calculation of the
scattering amplitude. We show that in the worldline formalism the 2-body
scattering amplitude written in impact parameter space naturally factorizes at
each loop order. This factorization is responsible for the exponentiation of
the eikonal phase, a result which was anticipated in the work of Mogull,
Plefka, and Steinhoff [2010.02865 [hep-th]].
| [
{
"version": "v1",
"created": "Thu, 19 Sep 2024 16:33:44 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:21:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Du",
"Yuchen",
""
],
[
"Ajith",
"Siddarth",
""
],
[
"Rajagopal",
"Ravisankar",
""
],
[
"Vaman",
"Diana",
""
]
] |
2409.14535 | Liangzhi Wang | Liangzhi Wang, Jie Zhang, Yuan Gao, Jiliang Zhang, Guiyi Wei, Haibo
Zhou, Bin Zhuge, and Zitian Zhang | Hyper-parameter Optimization for Wireless Network Traffic Prediction
Models with A Novel Meta-Learning Framework | This work has been submitted to the IEEE for possible publication | null | null | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes a novel meta-learning based hyper-parameter optimization
framework for wireless network traffic prediction (NTP) models. The primary
objective is to accumulate and leverage the acquired hyper-parameter
optimization experience, enabling the rapid determination of optimal
hyper-parameters for new tasks. In this paper, an attention-based deep neural
network (ADNN) is employed as the base-learner to address specific NTP tasks.
The meta-learner is an innovative framework that integrates meta-learning with
the k-nearest neighbor algorithm (KNN), genetic algorithm (GA), and gated
residual network (GRN). Specifically, KNN is utilized to identify a set of
candidate hyper-parameter selection strategies for a new task, which then
serves as the initial population for GA, while a GRN-based chromosome screening
module accelerates the validation of offspring chromosomes, ultimately
determining the optimal hyper-parameters. Experimental results demonstrate
that, compared to traditional methods such as Bayesian optimization (BO), GA,
and particle swarm optimization (PSO), the proposed framework determines
optimal hyper-parameters more rapidly, significantly reduces optimization time,
and enhances the performance of the base-learner. It achieves an optimal
balance between optimization efficiency and prediction accuracy.
| [
{
"version": "v1",
"created": "Sun, 22 Sep 2024 17:09:43 GMT"
},
{
"version": "v2",
"created": "Mon, 7 Apr 2025 17:54:41 GMT"
},
{
"version": "v3",
"created": "Sat, 3 May 2025 06:37:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wang",
"Liangzhi",
""
],
[
"Zhang",
"Jie",
""
],
[
"Gao",
"Yuan",
""
],
[
"Zhang",
"Jiliang",
""
],
[
"Wei",
"Guiyi",
""
],
[
"Zhou",
"Haibo",
""
],
[
"Zhuge",
"Bin",
""
],
[
"Zhang",
"Zitian",
""
]
] |
2409.14868 | Carmelo Cisto | Carmelo Cisto, Gioia Failla and Francesco Navarra | Generalized Numerical semigroups up to isomorphism | 18 pages | Communications in Algebra, 2025 | 10.1080/00927872.2025.2489604 | null | math.CO math.AC | http://creativecommons.org/licenses/by-nc-sa/4.0/ | A generalized numerical semigroup is a submonoid $S$ of $\mathbb{N}^d$ with
finite complement in it. We characterize isomorphisms between these monoids in
terms of permutation of coordinates. Considering the equivalence relation that
identifies the monoids obtained by the action of a permutation and establishing
a criterion to select a representative from each equivalence class, we define
some procedures for generating the set of all generalized numerical semigroups
of given genus up to isomorphism. Finally, we present computational data and
explore properties related to the number of generalized numerical semigroups of
a given genus up to isomorphism.
| [
{
"version": "v1",
"created": "Mon, 23 Sep 2024 10:05:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Cisto",
"Carmelo",
""
],
[
"Failla",
"Gioia",
""
],
[
"Navarra",
"Francesco",
""
]
] |
2409.15096 | Cody Stockdale | Cody B. Stockdale and Cody Waters | Compact pseudodifferential and Fourier integral operators via
localization | 19 pages | null | null | null | math.CA math.AP math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a general framework of localized operators, i.e., operators whose
matrix coefficients with respect to the Gabor frame are concentrated on the
diagonal. We show that localized operators are bounded between modulation
spaces, and we deduce their compactness from an easily verifiable weak
compactness condition. We apply this abstract formalism to unify and extend
existing theorems for pseudodifferential and Fourier integral operators, and to
obtain new results for three-parameter pseudodifferential operators.
| [
{
"version": "v1",
"created": "Mon, 23 Sep 2024 15:09:50 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 16:34:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Stockdale",
"Cody B.",
""
],
[
"Waters",
"Cody",
""
]
] |
2409.15193 | John Staunton | Ameya Chavda, Daniel Naegels, John Staunton | Fractonic Coset Construction for Spontaneously Broken Translations | 56 pages, 1 figure, 2 supplemental Mathematica notebooks | Phys.Rev.D 111 (2025) 8, 085018 | 10.1103/PhysRevD.111.085018 | null | hep-th cond-mat.other | http://creativecommons.org/licenses/by/4.0/ | We study the homogeneous breaking of spatial translation symmetry
concomitantly with the spontaneous breaking of other internal and spacetime
symmetries, including dilations. We use the symmetry-breaking pattern as the
only input to derive, via the coset construction, general effective field
theories for the symmetry-originated modes associated with Goldstone's theorem,
namely the Nambu-Goldstone candidates. Through explicit computations, we show
that integrating out the explicit massive Nambu-Goldstone candidates or
imposing symmetric constraints, namely the inverse Higgs constraints, to
express massive modes in terms of the massless ones leads to physically
distinct effective field theories. This sensitivity to the chosen method can be
traced back to the homogeneous breaking of translations, the homogeneous aspect
of the breaking induces a mixing between internal and spacetime symmetries at
the level of the Lie algebra. This, in turn, leads to subtle discussions about
the inverse Higgs constraints, in particular that they lead to a loss of
generality in our specific examples. The derived general effective field
theories also give rise to a broad class of theories exhibiting emergent
enhanced shift symmetries, which constrain the mobility of the modes. The
latter are referred to as fractonic modes.
| [
{
"version": "v1",
"created": "Mon, 23 Sep 2024 16:42:51 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 14:51:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chavda",
"Ameya",
""
],
[
"Naegels",
"Daniel",
""
],
[
"Staunton",
"John",
""
]
] |
2409.16023 | Anna Bobrikova | Anna Bobrikova, Juri Poutanen, Vladislav Loktev | Polarized radiation from the spreading layer of the weakly magnetized
neutron stars | 9 pages, 7 figures | A&A 696, A181 (2025) | 10.1051/0004-6361/202452358 | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | Observations show that the X-ray emission of the accreting weakly magnetized
neutron stars is polarized. Here, we develop a theoretical model, where we
assume the emission of the accreting neutron star coming from the spreading
layer, the extension of the boundary between the disk and the neutron star
surface onto the surface. We then calculate the Stokes parameters of the
emission accounting for relativistic aberration and gravitational light bending
in the Schwarzschild metric. We show that regardless of the geometry, for the
spreading layer, we cannot expect the polarization degree to be higher than
1.5%. Our results have implications with regard to the understanding of the
X-ray polarization from weakly magnetized neutron stars observed with the
Imaging X-ray Polarimetry Explorer and the future enhanced X-ray Timing and
Polarimetry mission.
| [
{
"version": "v1",
"created": "Tue, 24 Sep 2024 12:24:04 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:50:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bobrikova",
"Anna",
""
],
[
"Poutanen",
"Juri",
""
],
[
"Loktev",
"Vladislav",
""
]
] |
2409.16542 | Zhikai Zhou | Zhikai Zhou, Yin Kang, Weishi Wan and Chao Feng | Intense X-ray Vortices Generation via Wavefront Shaping in high-gain
Free-Electron Lasers | null | null | null | null | physics.optics physics.acc-ph | http://creativecommons.org/licenses/by/4.0/ | The x-ray vortex optical beam, distinguished by its topological charge and
orbital angular momentum, offers new insights in probing complex electronic
structures, enhancing material characterization, and advancing high-resolution
imaging techniques. Here we propose a straightforward and effective method to
generate intense x-ray vortices with tunable wavelengths in high-gain
free-electron lasers (FELs). By simply adjusting the wavefront tilt of the
radiation pulse during the FEL gain process, high-quality vortex beam can be
amplified until saturation. Compared to existing methods for FEL vortex
generation, the proposed technique imposes no wavelength limitations and can be
easily implemented in high-gain FEL facilities, regardless of the operation
modes.
| [
{
"version": "v1",
"created": "Wed, 25 Sep 2024 01:33:46 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 14:24:07 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhou",
"Zhikai",
""
],
[
"Kang",
"Yin",
""
],
[
"Wan",
"Weishi",
""
],
[
"Feng",
"Chao",
""
]
] |
2409.17447 | Andreas Hilfinger | Brayden Kell and Andreas Hilfinger | Efficiency-fluctuation trade-offs in biomolecular assembly processes | null | null | null | null | q-bio.MN q-bio.SC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stochastic fluctuations of molecular abundances are a ubiquitous feature of
cellular processes and lead to significant cell-to-cell variability. Recent
theoretical work established lower bounds for stochastic fluctuations in cells
for broad classes of cellular processes by analyzing the dynamics of reaction
motifs that are embedded within a larger network with arbitrary interactions
and dynamics. For example, a class of generalized assembly processes in which
two co-regulated subunits irreversibly form a complex was shown to exhibit an
unavoidable trade-off between assembly efficiency and subunit fluctuations:
Regardless of rate constants and details of feedback control, subunit
fluctuations were shown to diverge as the assembly efficiency approaches 100%.
In contrast, other work has reported how efficient assembly processes work as
stochastic noise filters or can achieve robust adaptation through integral
control. While all of these results are technically correct their seemingly
contradictory conclusions raise the question of how broadly applicable the
previously reported efficiency-fluctuation trade-off is. Here, we show that a
much broader class of assembly processes than previously considered is subject
to an efficiency-fluctuation trade-off which diverges in the high efficiency
regime. We find the proposed noise filtering property of efficient assembly
processes corresponds to a singular limit of this class of systems.
Additionally, we show that combining feedback control with distinct subunit
synthesis rates is a necessary condition to overcome the generalized
efficiency-fluctuation trade-off. Through numerical examples, we show that
biomolecular integral controllers are one of several realizations of such
control. How small a change to joint subunit control is sufficient to avoid
diverging fluctuations in the high-efficiency limit remains an open question.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 00:39:32 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 21:04:32 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Kell",
"Brayden",
""
],
[
"Hilfinger",
"Andreas",
""
]
] |
2409.17820 | Junghyun Shin | Junghyun Shin, Jae-Ho Han, Anjali Rathore, Joon Sue Lee, Seung-Bo
Shim, Jinwoong Cha, Sunghun Park, Junho Suh | Electric Control of Polarity in Spin-Orbit Josephson Diode | 39 pages, 18 figures, 1 table | null | null | null | cond-mat.supr-con cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The effect of spin-orbit coupling in a Josephson diode has not been
elucidated due to its interplay with the complexity of Josephson devices. Here,
we systematically control local electric fields in epitaxial Al-InAs Josephson
junctions under in-plane magnetic fields and observe a polarity reversal of the
Josephson diode. We interpret this polarity reversal as an effect of
field-tunable spin-orbit coupling on nonreciprocal Josephson currents. A
theoretical model, accounting for Rashba and Dresselhaus spin-orbit couplings
in a planar Josephson junction containing many transverse subbands, aligns with
the observed polarity reversal and its dependence on magnetic field. Our
finding addresses spin-orbit control in a Josephson diode, enabling
manipulation of Josephson harmonics.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 13:15:38 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 11:00:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Shin",
"Junghyun",
""
],
[
"Han",
"Jae-Ho",
""
],
[
"Rathore",
"Anjali",
""
],
[
"Lee",
"Joon Sue",
""
],
[
"Shim",
"Seung-Bo",
""
],
[
"Cha",
"Jinwoong",
""
],
[
"Park",
"Sunghun",
""
],
[
"Suh",
"Junho",
""
]
] |
2409.17866 | Diana Vaman | Siddarth Ajith, Yuchen Du, Ravisankar Rajagopal and Diana Vaman | Worldline Formalism, Eikonal Expansion and the Classical Limit of
Scattering Amplitudes | 75 pages, references added | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We revisit the fundamentals of two different methods for calculating
classical observables: the eikonal method, which is a scattering
amplitude-based method, and the worldline quantum field theory (WQFT) method.
The latter has been considered an extension of the worldline effective field
theory. We show that the eikonal and WQFT methods are equivalent and that
calculations can be translated freely between them.
Concretely, we focus on 2-into-2 scattering processes mediated by massless
force carriers. On the one hand, taking the classical limit of the QFT
scattering amplitude leads to the eikonal method. On the other hand, since in
the classical limit the scattering particles are almost on-shell throughout the
scattering process, the worldline, a first quantized formalism, is the most
efficient framework to study the scattering amplitude. This is an alternate but
equivalent formalism to the quantum field theoretic (QFT) framework. By taking
the classical limit of the scattering amplitude computed in the worldline, we
can derive the WQFT rules of Mogull, Plefka and Steinhoff. In WQFT, the Feynman
diagrams are reorganized into a new set of diagrams that facilitate the $\hbar$
expansion. Unlike the QFT eikonal method, which works recursively in
identifying the eikonal phase, the worldline-based computation allows to target
and systematically extract the classical contributions directly through a
specific set of WQFT diagrams. In worldline formalism the perturbative
expansion of the scattering amplitude is naturally organized in diagrams which
factorize (reducible) and diagrams which are new to that order (irreducible),
in a one-to-one map with the structure of the amplitude in the eikonal method.
This opened up the possibility to investigate and prove the conjectured
exponentiation of the eikonal phase in arXiv: 2409.12895.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 14:16:22 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:13:51 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ajith",
"Siddarth",
""
],
[
"Du",
"Yuchen",
""
],
[
"Rajagopal",
"Ravisankar",
""
],
[
"Vaman",
"Diana",
""
]
] |
2409.18167 | Zhaoyi Li | Zhaoyi Li, Honghao Fu, Takuya Isogawa, Isaac Chuang | Optimal Quantum Purity Amplification | 8+22 pages, 5+8 figures, 0+3 tables. Comments are welcome! | null | null | MIT-CTP/5775 | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum purity amplification (QPA) provides a novel approach to counteracting
the pervasive noise that degrades quantum states. We present the optimal QPA
protocol for general quantum systems and global noise, resolving a two-decade
open problem. Under strong depolarization, our protocol achieves an exponential
reduction in sample complexity over the best-known methods. We provide an
efficient implementation of the protocol based on generalized quantum phase
estimation. Additionally, we introduce SWAPNET, a sparse and shallow circuit
that enables QPA for near-term experiments. Numerical simulations demonstrate
the effectiveness of our protocol applied to quantum simulation of Hamiltonian
evolution, enhancing the fidelity of input states even under circuit-level
noise. Our findings suggest that QPA could improve the performance of quantum
information processing tasks, particularly in the context of Noisy
Intermediate-Scale Quantum (NISQ) devices, where reducing the effect of noise
with limited resources is critical.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 17:46:00 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Mar 2025 18:04:39 GMT"
},
{
"version": "v3",
"created": "Sat, 3 May 2025 02:55:06 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Li",
"Zhaoyi",
""
],
[
"Fu",
"Honghao",
""
],
[
"Isogawa",
"Takuya",
""
],
[
"Chuang",
"Isaac",
""
]
] |
2409.18262 | Evan McKinney | Evan McKinney, Israa G. Yusuf, Girgis Falstin, Gaurav Agarwal, Michael
Hatridge, Alex K. Jones | Spectator-Aware Frequency Allocation in Tunable-Coupler Quantum
Architectures | 10 pages, 12 figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper addresses frequency crowding in SNAIL-based superconducting
quantum modules. First, we present design constraints by describing a physical
model for realizable gates within a module, and building a fidelity model using
error budgeting derived from device characteristics. Second, we tackle the
allocation problem by analyzing the impact of frequency crowding on gate
fidelity as the radix of the module increases. We explore whether the heuristic
gate fidelity can be optimized with a discrete set of qubit frequencies while
adhering to defined separation thresholds. By leveraging a combination of
analytical and numerical techniques, we demonstrate scalable frequency
allocation strategies that minimize spectator-induced errors. Our results
further indicate that removing edges leads to improved gate fidelities while
maintaining sufficient connectivity, suggesting that edge density is not a
limiting factor for NISQ-scale benchmarks. The findings have implications for
designing robust, high-fidelity quantum systems with practical constraints on
hardware and connectivity.
| [
{
"version": "v1",
"created": "Thu, 26 Sep 2024 20:15:45 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Sep 2024 14:05:51 GMT"
},
{
"version": "v3",
"created": "Wed, 9 Oct 2024 20:04:12 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 15:27:34 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"McKinney",
"Evan",
""
],
[
"Yusuf",
"Israa G.",
""
],
[
"Falstin",
"Girgis",
""
],
[
"Agarwal",
"Gaurav",
""
],
[
"Hatridge",
"Michael",
""
],
[
"Jones",
"Alex K.",
""
]
] |
2409.19950 | Faranak Farshadifar | Faranak Farshadifar | Nil-prime ideals of a commutative ring | null | null | null | null | math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let R be a commutative ring with identity and N(R) be the set of all
nilpotent elements of R. The aim of this paper is to introduce and study the
notion of nil-prime ideals as a generalization of prime ideals. We say that a
proper ideal P of R is a nil-prime ideal if there exists x \in N(R) and
whenever ab \in P, then a \in P or b \in P or a+x \in P or b+x \in P for each
a,b \in R. Also, we introduce nil versions of some algebraic concepts in ring
theory such as nil-maximal ideal, nil-minimal ideal, nil-principal ideal and
investigate some nil-version of a well-known results about them.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 05:10:16 GMT"
},
{
"version": "v2",
"created": "Sun, 9 Feb 2025 09:26:29 GMT"
},
{
"version": "v3",
"created": "Mon, 24 Feb 2025 04:58:02 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 15:29:39 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Farshadifar",
"Faranak",
""
]
] |
2409.20254 | Maciej Grzeskowiak | Maciej Grze\'skowiak | MNT Elliptic Curves with Non-Prime Order | null | null | null | null | cs.CR math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Miyaji, Nakabayashi, and Takano proposed the algorithm for the construction
of prime order pairing-friendly elliptic curves with embedding degrees
$k=3,4,6$. We present a method for generating generalized MNT curves. The order
of such pairing-friendly curves is the product of two prime numbers.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 12:43:31 GMT"
},
{
"version": "v2",
"created": "Mon, 31 Mar 2025 17:10:42 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 16:20:21 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Grześkowiak",
"Maciej",
""
]
] |
2409.20281 | Mikko Korhonen | Mikko Korhonen | Structure of an exotic $2$-local subgroup in $E_7(q)$ | 9 pages | null | null | null | math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $G$ be the finite simple group of Lie type $G = E_7(q)$, where $q$ is an
odd prime power. Then $G$ is an index $2$ subgroup of the adjoint group
$G_{\operatorname{ad}}$, which is also denoted by $G_{\operatorname{ad}} =
\operatorname{Inndiag}(G)$ and known as the group of inner-diagonal
automorphisms. It was proven by Cohen--Liebeck--Saxl--Seitz (1992) that there
is an elementary abelian $2$-subgroup $E$ of order $4$ in
$G_{\operatorname{ad}}$, such that $N_{G_{\operatorname{ad}}}(E)/C_{G_{ad}}(E)
\cong \operatorname{Sym}_3$, and $C_{G_{\operatorname{ad}}}(E) = E \times
\operatorname{Inndiag}(D_4(q))$. Furthermore, such an $E$ is unique up to
conjugacy in $G_{\operatorname{ad}}$.
It is known that $N_G(E)$ is always a maximal subgroup of $G$, and
$N_{G_{\operatorname{ad}}}(E)$ is a maximal subgroup of $G_{\operatorname{ad}}$
unless $N_{G_{\operatorname{ad}}}(E) \leq G$. In this note, we describe the
structure of $N_{G}(E)$. It turns out that $N_G(E) =
N_{G_{\operatorname{ad}}}(E)$ if and only if $q \equiv \pm 1 \mod{8}$.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 13:36:54 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 05:04:10 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Korhonen",
"Mikko",
""
]
] |
2409.20379 | Chiara Moretti | Chiara Moretti, Maximilian Autenrieth, Riccardo Serra, Roberto Trotta,
David A. van Dyk and Andrei Mesinger | StratLearn-z: Improved photo-$z$ estimation from spectroscopic data
subject to selection effects | Accepted for publication on the Open Journal of Astrophysics | null | 10.33232/001c.137525 | null | astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | A precise measurement of photometric redshifts (photo-z) is key for the
success of modern photometric galaxy surveys. Machine learning (ML) methods
show great promise in this context, but suffer from covariate shift (CS) in
training sets due to selection bias where interesting sources are
underrepresented, and the corresponding ML models show poor generalisation
properties. We present an application of the StratLearn method to the
estimation of photo-z, validating against simulations where we enforce the
presence of CS to different degrees. StratLearn is a statistically principled
approach that relies on splitting the source and target datasets into strata
based on estimated propensity scores (i.e. the probability for an object to be
in the source set given its observed covariates). After stratification, two
conditional density estimators are fit separately to each stratum, then
combined via a weighted average. We benchmark our results against the GPz
algorithm, quantifying the performance of the two codes with a set of metrics.
Our results show that the StratLearn-z metrics are only marginally affected by
the presence of CS, while GPz shows a significant degradation of performance in
the photo-z prediction for fainter objects. For the strongest CS scenario,
StratLearn-z yields a reduced fraction of catastrophic errors, a factor of 2
improvement for the RMSE and one order of magnitude improvement on the bias. We
also assess the quality of the conditional redshift estimates with the
probability integral transform (PIT). The PIT distribution obtained from
StratLearn-z features fat fewer outliers and is symmetric, i.e. the predictions
appear to be centered around the true redshift value, despite showing a
conservative estimation of the spread of the conditional redshift
distributions. Our julia implementation of the method is available at
https://github.com/chiaramoretti/StratLearn-z.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 15:14:17 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 14:09:06 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Moretti",
"Chiara",
""
],
[
"Autenrieth",
"Maximilian",
""
],
[
"Serra",
"Riccardo",
""
],
[
"Trotta",
"Roberto",
""
],
[
"van Dyk",
"David A.",
""
],
[
"Mesinger",
"Andrei",
""
]
] |
2409.20538 | Henning Schl\"omer | Henning Schl\"omer and Subir Sachdev | Quantum Annealing with chaotic driver Hamiltonians | 9+2 pages | Annals of Physics 479, 170042 (2025) | 10.1016/j.aop.2025.170042 | null | quant-ph cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum annealing is a computational approach designed to leverage quantum
fluctuations for solving large-scale classical optimization problems. Although
incorporating standard transverse field (TF) terms in the annealing process can
help navigate sharp minima, the potential for achieving a scalable quantum
advantage for general optimization problems remains uncertain. Here, we examine
the effectiveness of including chaotic quantum driver Hamiltonians in the
annealing dynamics. Specifically, we investigate driver Hamiltonians based on a
bosonic spin version of the Sachdev-Ye-Kitaev (SYK) model, which features a
high degree of non-locality and non-commutativity. Focusing on MaxCut instances
on regular graphs, we find that a considerable proportion of SYK model
instances demonstrate significant speedups, especially for challenging graph
configurations. Additionally, our analysis of time-to-solution scalings for the
low autocorrelation binary sequence (LABS) problem suggests that SYK-type
fluctuations can outperform traditional transverse field annealing schedules in
large-scale optimization tasks.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 17:39:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Schlömer",
"Henning",
""
],
[
"Sachdev",
"Subir",
""
]
] |
2410.00261 | Kejia Ren | Kejia Ren, Gaotian Wang, Andrew S. Morgan, Lydia E. Kavraki, and Kaiyu
Hang | Object-Centric Kinodynamic Planning for Nonprehensile Robot
Rearrangement Manipulation | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Nonprehensile actions such as pushing are crucial for addressing multi-object
rearrangement problems. To date, existing nonprehensile solutions are all
robot-centric, i.e., the manipulation actions are generated with robot-relevant
intent and their outcomes are passively evaluated afterwards. Such pipelines
are very different from human strategies and are typically inefficient. To this
end, this work proposes a novel object-centric planning paradigm and develops
the first object-centric planner for general nonprehensile rearrangement
problems. By assuming that each object can actively move without being driven
by robot interactions, the object-centric planner focuses on planning desired
object motions, which are realized via robot actions generated online via a
closed-loop pushing strategy. Through extensive experiments and in comparison
with state-of-the-art baselines in both simulation and on a physical robot, we
show that our object-centric paradigm can generate more intuitive and
task-effective robot actions with significantly improved efficiency. In
addition, we propose a benchmarking protocol to standardize and facilitate
future research in nonprehensile rearrangement.
| [
{
"version": "v1",
"created": "Mon, 30 Sep 2024 22:16:51 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 01:56:56 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ren",
"Kejia",
""
],
[
"Wang",
"Gaotian",
""
],
[
"Morgan",
"Andrew S.",
""
],
[
"Kavraki",
"Lydia E.",
""
],
[
"Hang",
"Kaiyu",
""
]
] |
2410.00543 | Li-Gang Zhu | Li-Gang Zhu, Guoyang Fu, Shulan Li, Dan Zhang, Jian-Pin Wu | Quasinormal modes of a charged loop quantum black hole | null | Phys.Rev.D 111 (2025) 10, 104008 | 10.1103/PhysRevD.111.104008 | null | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study presents a systematic investigation of quasinormal modes (QNMs)
for probe fields-massless/massive scalar and Dirac fields-around a charged loop
quantum gravity black hole (LQG-BH) characterized by the quantum parameter
$b_0$ and the charge parameter $Q$. Through spectral analysis of quasinormal
frequencies (QNFs), we uncover a distinct overtone outburst driven by quantum
gravity effects, prominently manifested in the scalar field spectrum with the
multipole quantum number $l=0$. Both the outburst and its accompanying
oscillatory patterns grow more pronounced with increasing overtone numbers. In
contrast, massless scalar fields with $l>1$ and Dirac fields exhibit delayed
outburst development, with non-monotonic behavior dominating the first two
overtones. Notably, increasing the charge $Q$ universally suppresses
quantum-gravity-induced features, including outbursts, non-monotonicity, and
oscillations. Furthermore, we present evidence suggesting the presence of
quasi-resonances in the massive scalar QNM spectrum, thereby illustrating the
potential for the emergence of arbitrarily long-lived modes in this charged LQG
spacetime. These findings establish a robust and universal interplay between
quantum gravity effects and charge dynamics, providing new insights into the
spectral properties of quantum-corrected BHs.
| [
{
"version": "v1",
"created": "Tue, 1 Oct 2024 09:35:10 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Oct 2024 12:58:19 GMT"
},
{
"version": "v3",
"created": "Sun, 4 May 2025 03:18:28 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhu",
"Li-Gang",
""
],
[
"Fu",
"Guoyang",
""
],
[
"Li",
"Shulan",
""
],
[
"Zhang",
"Dan",
""
],
[
"Wu",
"Jian-Pin",
""
]
] |
2410.01250 | Shaozu Ding | Shaozu Ding, Yihong Tang, Marco De Vincenzi, Dajiang Suo | High and Low Resolution Tradeoffs in Roadside Multimodal Sensing | 8 pages, 7 figures | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Balancing cost and performance is crucial when choosing high- versus
low-resolution point-cloud roadside sensors. For example, LiDAR delivers dense
point cloud, while 4D millimeter-wave radar, though spatially sparser, embeds
velocity cues that help distinguish objects and come at a lower price.
Unfortunately, the sensor placement strategies will influence point cloud
density and distribution across the coverage area. Compounding the first
challenge is the fact that different sensor mixtures often demand distinct
neural network architectures to maximize their complementary strengths. Without
an evaluation framework that establishes a benchmark for comparison, it is
imprudent to make claims regarding whether marginal gains result from higher
resolution and new sensing modalities or from the algorithms. We present an
ex-ante evaluation that addresses the two challenges. First, we realized a
simulation tool that builds on integer programming to automatically compare
different sensor placement strategies against coverage and cost jointly.
Additionally, inspired by human multi-sensory integration, we propose a modular
framework to assess whether reductions in spatial resolution can be compensated
by informational richness in detecting traffic participants. Extensive
experimental testing on the proposed framework shows that fusing
velocity-encoded radar with low-resolution LiDAR yields marked gains (14
percent AP for pedestrians and an overall mAP improvement of 1.5 percent across
six categories) at lower cost than high-resolution LiDAR alone. Notably, these
marked gains hold regardless of the specific deep neural modules employed in
our frame. The result challenges the prevailing assumption that high resolution
are always superior to low-resolution alternatives.
| [
{
"version": "v1",
"created": "Wed, 2 Oct 2024 05:49:34 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 14:03:44 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ding",
"Shaozu",
""
],
[
"Tang",
"Yihong",
""
],
[
"De Vincenzi",
"Marco",
""
],
[
"Suo",
"Dajiang",
""
]
] |
2410.01435 | Giacomo Santoni | Giacomo Santoni, Francesco Scardino | Superfield twist-$2$ operators in $\mathcal{N} = 1$ SCFTs and their
renormalization-group improved generating functional in $\mathcal{N} = 1$ SYM
theory | 96 pages | JHEP 04 (2025) 159 | 10.1007/JHEP04(2025)159 | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We provide a new construction of superfield collinear twist-$2$ operators as
infinite-dimensional, irreducible representations of the collinear
superconformal algebra in $\mathcal{N}=1$ superconformal field theories. As an
application, we realize the above representations in terms of free superfields,
in a manifestly gauge-invariant and supersymmetric-covariant fashion, in the
zero coupling limit of $\mathcal{N}=1$ supersymmetric Yang-Mills (SYM) theory.
This realization makes manifest their mixing and renormalization properties at
one loop. We also extend to the superfield formalism the perturbative and
nonperturbative techniques in [1-7] to a large class of supersymmetric theories
that are superconformal in the zero-coupling limit. Specifically, we compute
the generating functional of superfield twist-$2$ operators in $\mathcal{N}=1$
SU($N$) SYM theory in the zero coupling limit. We also work out in a closed
form the corresponding asymptotic renormalization-group improved generating
functional in Euclidean superspace and its planar and leading nonplanar
large-$N$ expansion. We verify -- as originally predicted in [5] and verified
in the component formalism [3, 4, 6, 7] -- that the leading nonplanar
asymptotic RG-improved generating functional matches the structure of logarithm
of a functional superdeterminant of the corresponding nonperturbative object,
which it should be asymptotic to at short distances because of the asymptotic
freedom. Hence, our large-$N$ computation sets strong ultraviolet asymptotic
constraints on the nonperturbative solution of large-$N$ $\mathcal{N} = 1$ SYM
theory that may be a pivotal guide for the search of such a solution.
| [
{
"version": "v1",
"created": "Wed, 2 Oct 2024 11:37:49 GMT"
},
{
"version": "v2",
"created": "Mon, 7 Oct 2024 11:01:01 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Mar 2025 09:42:19 GMT"
},
{
"version": "v4",
"created": "Sun, 4 May 2025 21:32:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Santoni",
"Giacomo",
""
],
[
"Scardino",
"Francesco",
""
]
] |
2410.01566 | Zhiyu Liu | Yuchen Liu, Zhiyu Liu, Chenyang Xu | Irreducible symplectic varieties with a large second Betti number | 28 pages. Comments are welcome! v5: A footnote is added since
Koll\'ar pointed out that our example is not smooth (see footnote 2). v4: To
appear in Crelle's Journal | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove a general result on the existence of irreducible symplectic
compactifications of non-compact Lagrangian fibrations. As an application, we
show that the relative Jacobian fibration of cubic fivefolds containing a fixed
cubic fourfold can be compactified by a $\mathbb{Q}$-factorial terminal
irreducible symplectic variety with the second Betti number at least 24, and
admits a Lagrangian fibration whose base is a weighted projective space. In
particular, it belongs to a new deformation type of irreducible symplectic
varieties.
| [
{
"version": "v1",
"created": "Wed, 2 Oct 2024 14:03:48 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Oct 2024 16:07:04 GMT"
},
{
"version": "v3",
"created": "Mon, 3 Feb 2025 02:24:17 GMT"
},
{
"version": "v4",
"created": "Tue, 1 Apr 2025 14:41:54 GMT"
},
{
"version": "v5",
"created": "Fri, 2 May 2025 22:02:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liu",
"Yuchen",
""
],
[
"Liu",
"Zhiyu",
""
],
[
"Xu",
"Chenyang",
""
]
] |
2410.01859 | Yubo Li | Yubo Li, Rema Padman | Enhancing End Stage Renal Disease Outcome Prediction: A Multi-Sourced
Data-Driven Approach | null | null | null | null | q-bio.QM cs.LG | http://creativecommons.org/licenses/by/4.0/ | Objective: To improve prediction of Chronic Kidney Disease (CKD) progression
to End Stage Renal Disease (ESRD) using machine learning (ML) and deep learning
(DL) models applied to an integrated clinical and claims dataset of varying
observation windows, supported by explainable AI (XAI) to enhance
interpretability and reduce bias.
Materials and Methods: We utilized data about 10,326 CKD patients, combining
their clinical and claims information from 2009 to 2018. Following data
preprocessing, cohort identification, and feature engineering, we evaluated
multiple statistical, ML and DL models using data extracted from five distinct
observation windows. Feature importance and Shapley value analysis were
employed to understand key predictors. Models were tested for robustness,
clinical relevance, misclassification errors and bias issues.
Results: Integrated data models outperformed those using single data sources,
with the Long Short-Term Memory (LSTM) model achieving the highest AUC (0.93)
and F1 score (0.65). A 24-month observation window was identified as optimal
for balancing early detection and prediction accuracy. The 2021 eGFR equation
improved prediction accuracy and reduced racial bias, notably for African
American patients. Discussion: Improved ESRD prediction accuracy, results
interpretability and bias mitigation strategies presented in this study have
the potential to significantly enhance CKD and ESRD management, support
targeted early interventions and reduce healthcare disparities.
Conclusion: This study presents a robust framework for predicting ESRD
outcomes in CKD patients, improving clinical decision-making and patient care
through multi-sourced, integrated data and AI/ML methods. Future research will
expand data integration and explore the application of this framework to other
chronic diseases.
| [
{
"version": "v1",
"created": "Wed, 2 Oct 2024 03:21:01 GMT"
},
{
"version": "v2",
"created": "Tue, 8 Oct 2024 18:54:06 GMT"
},
{
"version": "v3",
"created": "Mon, 20 Jan 2025 20:01:56 GMT"
},
{
"version": "v4",
"created": "Sat, 3 May 2025 03:34:49 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Li",
"Yubo",
""
],
[
"Padman",
"Rema",
""
]
] |
2410.02597 | Hainan Xu | Hainan Xu, Travis M. Bartley, Vladimir Bataev, Boris Ginsburg | HAINAN: Fast and Accurate Transducer for Hybrid-Autoregressive ASR | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present Hybrid-Autoregressive INference TrANsducers (HAINAN), a novel
architecture for speech recognition that extends the Token-and-Duration
Transducer (TDT) model. Trained with randomly masked predictor network outputs,
HAINAN supports both autoregressive inference with all network components and
non-autoregressive inference without the predictor. Additionally, we propose a
novel semi-autoregressive inference paradigm that first generates an initial
hypothesis using non-autoregressive inference, followed by refinement steps
where each token prediction is regenerated using parallelized autoregression on
the initial hypothesis. Experiments on multiple datasets across different
languages demonstrate that HAINAN achieves efficiency parity with CTC in
non-autoregressive mode and with TDT in autoregressive mode. In terms of
accuracy, autoregressive HAINAN outperforms TDT and RNN-T, while
non-autoregressive HAINAN significantly outperforms CTC. Semi-autoregressive
inference further enhances the model's accuracy with minimal computational
overhead, and even outperforms TDT results in some cases. These results
highlight HAINAN's flexibility in balancing accuracy and speed, positioning it
as a strong candidate for real-world speech recognition applications.
| [
{
"version": "v1",
"created": "Thu, 3 Oct 2024 15:38:20 GMT"
},
{
"version": "v2",
"created": "Mon, 24 Feb 2025 18:15:09 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 08:20:01 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Xu",
"Hainan",
""
],
[
"Bartley",
"Travis M.",
""
],
[
"Bataev",
"Vladimir",
""
],
[
"Ginsburg",
"Boris",
""
]
] |
2410.03342 | Shantanu Desai | Rayani Venkat Sai Rithvik, Shantanu Desai | A meta-analysis of impact factors of astrophysics journals | 10 pages, 2 figures. More journals added. Also added miscellaneous
publication statistics as well as APC details. Accepted for publication in
EPJP | null | null | null | astro-ph.IM cs.DL physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | We calculate the 2024 impact factors for the 38 most widely used journals in
Astrophysics, using the citations collated by NASA/ADS (Astrophysics Data
System) and compare them to the official impact factors. This includes journals
which publish papers outside of astrophysics such as PRD, EPJC, Nature, etc. We
also propose a new metric to gauge the impact factor based on the median number
of citations in a journal and calculate the same for all the journals. We find
that the ADS-based impact factors are mostly in agreement, albeit higher than
the official impact factors for most journals. The journals with the maximum
fractional difference in median-based and old impact factors are JHEAP and
PTEP. We find the maximum difference between the ADS and official impact factor
for Nature.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 11:58:06 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 16:56:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Rithvik",
"Rayani Venkat Sai",
""
],
[
"Desai",
"Shantanu",
""
]
] |
2410.03613 | Jie Xiao | Jie Xiao, Qianyi Huang, Xu Chen and Chen Tian | Understanding Large Language Models in Your Pockets: Performance Study
on COTS Mobile Devices | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As large language models (LLMs) increasingly integrate into every aspect of
our work and daily lives, there are growing concerns about user privacy, which
push the trend toward local deployment of these models. There are a number of
lightweight LLMs (e.g., Gemini Nano, LLAMA2 7B) that can run locally on
smartphones, providing users with greater control over their personal data. As
a rapidly emerging application, we are concerned about their performance on
commercial-off-the-shelf mobile devices. To fully understand the current
landscape of LLM deployment on mobile platforms, we conduct a comprehensive
measurement study on mobile devices. We evaluate both metrics that affect user
experience, including token throughput, latency, and battery consumption, as
well as factors critical to developers, such as resource utilization, DVFS
strategies, and inference engines. In addition, we provide a detailed analysis
of how these hardware capabilities and system dynamics affect on-device LLM
performance, which may help developers identify and address bottlenecks for
mobile LLM applications. We also provide comprehensive comparisons across the
mobile system-on-chips (SoCs) from major vendors, highlighting their
performance differences in handling LLM workloads. We hope that this study can
provide insights for both the development of on-device LLMs and the design for
future mobile system architecture.
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 17:14:59 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 13:34:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Xiao",
"Jie",
""
],
[
"Huang",
"Qianyi",
""
],
[
"Chen",
"Xu",
""
],
[
"Tian",
"Chen",
""
]
] |
2410.03954 | Amir Eskandari | Amir Eskandari, Aman Anand, Drishti Sharma, Farhana Zulkernine | SDA-GRIN for Adaptive Spatial-Temporal Multivariate Time Series
Imputation | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In various applications, the multivariate time series often suffers from
missing data. This issue can significantly disrupt systems that rely on the
data. Spatial and temporal dependencies can be leveraged to impute the missing
samples. Existing imputation methods often ignore dynamic changes in spatial
dependencies. We propose a Spatial Dynamic Aware Graph Recurrent Imputation
Network (SDA-GRIN) which is capable of capturing dynamic changes in spatial
dependencies.SDA-GRIN leverages a multi-head attention mechanism to adapt graph
structures with time. SDA-GRIN models multivariate time series as a sequence of
temporal graphs and uses a recurrent message-passing architecture for
imputation. We evaluate SDA-GRIN on four real-world datasets: SDA-GRIN improves
MSE by 9.51% for the AQI and 9.40% for AQI-36. On the PEMS-BAY dataset, it
achieves a 1.94% improvement in MSE. Detailed ablation study demonstrates the
effect of window sizes and missing data on the performance of the method.
Project page:https://ameskandari.github.io/sda-grin/
| [
{
"version": "v1",
"created": "Fri, 4 Oct 2024 22:32:08 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 15:55:16 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Eskandari",
"Amir",
""
],
[
"Anand",
"Aman",
""
],
[
"Sharma",
"Drishti",
""
],
[
"Zulkernine",
"Farhana",
""
]
] |
2410.05016 | Hugo Thimonier | Hugo Thimonier, Jos\'e Lucas De Melo Costa, Fabrice Popineau, Arpad
Rimmel, Bich-Li\^en Doan | T-JEPA: Augmentation-Free Self-Supervised Learning for Tabular Data | Accepted at ICLR 2025: https://openreview.net/forum?id=gx3LMRB15C | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Self-supervision is often used for pre-training to foster performance on a
downstream task by constructing meaningful representations of samples.
Self-supervised learning (SSL) generally involves generating different views of
the same sample and thus requires data augmentations that are challenging to
construct for tabular data. This constitutes one of the main challenges of
self-supervision for structured data. In the present work, we propose a novel
augmentation-free SSL method for tabular data. Our approach, T-JEPA, relies on
a Joint Embedding Predictive Architecture (JEPA) and is akin to mask
reconstruction in the latent space. It involves predicting the latent
representation of one subset of features from the latent representation of a
different subset within the same sample, thereby learning rich representations
without augmentations. We use our method as a pre-training technique and train
several deep classifiers on the obtained representation. Our experimental
results demonstrate a substantial improvement in both classification and
regression tasks, outperforming models trained directly on samples in their
original data space. Moreover, T-JEPA enables some methods to consistently
outperform or match the performance of traditional methods likes Gradient
Boosted Decision Trees. To understand why, we extensively characterize the
obtained representations and show that T-JEPA effectively identifies relevant
features for downstream tasks without access to the labels. Additionally, we
introduce regularization tokens, a novel regularization method critical for
training of JEPA-based models on structured data.
| [
{
"version": "v1",
"created": "Mon, 7 Oct 2024 13:15:07 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Dec 2024 09:49:25 GMT"
},
{
"version": "v3",
"created": "Sat, 3 May 2025 06:48:53 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Thimonier",
"Hugo",
""
],
[
"Costa",
"José Lucas De Melo",
""
],
[
"Popineau",
"Fabrice",
""
],
[
"Rimmel",
"Arpad",
""
],
[
"Doan",
"Bich-Liên",
""
]
] |
2410.05389 | Caio Bastos de Senna Nascimento | Caio Nascimento and Marilena Loverde | Cosmological perturbation theory for large scale structure in phase
space | 21+15 pages, 9 figures. Matching JCAP accepted version | null | null | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We develop a framework for Large Scale Structure (LSS) perturbation theory,
that solves the Vlasov-Poisson system of equations for the distribution
function in full phase space. This approach relaxes the usual apriori
assumption of negligible velocity dispersion underlying the Standard
Perturbation Theory (SPT). We apply the new method to rederive the usual SPT
kernels up to third order in the perturbative expansion. We also show that a
counterterm, identical to the one introduced by standard Effective Field Theory
(EFT) methods, naturally arises within our framework. We finish by making a
precise connection to EFT techniques, which reveals the necessity of the
EFTofLSS to self-consistently model the long-wavelength fluid, and illustrates
the importance of having theoretical control over short distance fluctuations.
| [
{
"version": "v1",
"created": "Mon, 7 Oct 2024 18:00:08 GMT"
},
{
"version": "v2",
"created": "Mon, 2 Dec 2024 01:14:01 GMT"
},
{
"version": "v3",
"created": "Sun, 4 May 2025 21:00:32 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Nascimento",
"Caio",
""
],
[
"Loverde",
"Marilena",
""
]
] |
2410.06525 | Haoran Guan Mr | Haoran Guan, Yuwei Fan | Shifted CholeskyQR for sparse matrices | null | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we focus on Shifted CholeskyQR for sparse matrices, which is
widely used in real applications. We introduce a new model for sparse matrices,
categorizing them into two types, $T_{1}$ matrices and $T_{2}$ matrices, based
on the presence of dense columns. We provide an alternative choice of the
shifted item $s$ for Shifted CholeskyQR3 \cite{Shifted} based on the structure
and the key element of the input sparse $X$. We do rounding error analysis of
Shifted CholeskyQR3 with such an $s$ and show that it is optimal compared to
the original one in \cite{New} with proper element-norm conditions (ENCs) for
$T_{1}$ matrices, improving the applicability while maintaining numerical
stability. Our theoretical analysis utilizes the properties of $[\cdot]_{g}$
proposed in \cite{New}, which is the first to build connections between
rounding error analysis and sparse matrices. Numerical experiments confirm our
findings for $T_{1}$ matrices. Additionally, Shifted CholeskyQR3 with the
alternative choice $s$ is applicable to $T_{2}$ matrices, which are more
ill-conditioned than dense cases. Furthermore, Shifted CholeskyQR3 with our
alternative $s$ shows good efficiency for both $T_{1}$ and $T_{2}$ matrices.
| [
{
"version": "v1",
"created": "Wed, 9 Oct 2024 03:56:03 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Feb 2025 10:37:06 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 09:23:35 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Guan",
"Haoran",
""
],
[
"Fan",
"Yuwei",
""
]
] |
2410.07414 | Tao Zhang | Tao Zhang, Rajagopal Venkatesaramani, Rajat K. De, Bradley A. Malin,
Yevgeniy Vorobeychik | Bayes-Nash Generative Privacy Against Membership Inference Attacks | arXiv admin note: substantial text overlap with arXiv:2406.01811 | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | Membership inference attacks (MIAs) expose significant privacy risks by
determining whether an individual's data is in a dataset. While differential
privacy (DP) mitigates such risks, it has several limitations in achieving an
optimal balance between utility and privacy, include limited resolution in
expressing this tradeoff in only a few privacy parameters, and intractable
sensitivity calculations that may be necessary to provide tight privacy
guarantees. We propose a game-theoretic framework that models privacy
protection from MIA as a Bayesian game between a defender and an attacker. In
this game, a dataset is the defender's private information, with privacy loss
to the defender (which is gain to the attacker) captured in terms of the
attacker's ability to infer membership of individuals in the dataset. To
address the strategic complexity of this game, we represent the mixed strategy
of the defender as a neural network generator which maps a private dataset to
its public representation (for example, noisy summary statistics), while the
mixed strategy of the attacker is captured by a discriminator which makes
membership inference claims. We refer to the resulting computational approach
as a general-sum Generative Adversarial Network, which is trained iteratively
by alternating generator and discriminator updates akin to conventional GANs.
We call the defender's data sharing policy thereby obtained Bayes-Nash
Generative Privacy (BNGP). The BNGP strategy avoids sensitivity calculations,
supports compositions of correlated mechanisms, is robust to the attacker's
heterogeneous preferences over true and false positives, and yields provable
differential privacy guarantees, albeit in an idealized setting.
| [
{
"version": "v1",
"created": "Wed, 9 Oct 2024 20:29:04 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Dec 2024 17:14:04 GMT"
},
{
"version": "v3",
"created": "Thu, 13 Feb 2025 17:27:11 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 02:00:50 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhang",
"Tao",
""
],
[
"Venkatesaramani",
"Rajagopal",
""
],
[
"De",
"Rajat K.",
""
],
[
"Malin",
"Bradley A.",
""
],
[
"Vorobeychik",
"Yevgeniy",
""
]
] |
2410.07577 | Benjamin Planche | Qucheng Peng, Benjamin Planche, Zhongpai Gao, Meng Zheng, Anwesa
Choudhuri, Terrence Chen, Chen Chen, Ziyan Wu | 3D Vision-Language Gaussian Splatting | Accepted at ICLR 2025. Main paper + supplementary material | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in 3D reconstruction methods and vision-language models
have propelled the development of multi-modal 3D scene understanding, which has
vital applications in robotics, autonomous driving, and virtual/augmented
reality. However, current multi-modal scene understanding approaches have
naively embedded semantic representations into 3D reconstruction methods
without striking a balance between visual and language modalities, which leads
to unsatisfying semantic rasterization of translucent or reflective objects, as
well as over-fitting on color modality. To alleviate these limitations, we
propose a solution that adequately handles the distinct visual and semantic
modalities, i.e., a 3D vision-language Gaussian splatting model for scene
understanding, to put emphasis on the representation learning of language
modality. We propose a novel cross-modal rasterizer, using modality fusion
along with a smoothed semantic indicator for enhancing semantic rasterization.
We also employ a camera-view blending technique to improve semantic consistency
between existing and synthesized views, thereby effectively mitigating
over-fitting. Extensive experiments demonstrate that our method achieves
state-of-the-art performance in open-vocabulary semantic segmentation,
surpassing existing methods by a significant margin.
| [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 03:28:29 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 17:00:03 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Peng",
"Qucheng",
""
],
[
"Planche",
"Benjamin",
""
],
[
"Gao",
"Zhongpai",
""
],
[
"Zheng",
"Meng",
""
],
[
"Choudhuri",
"Anwesa",
""
],
[
"Chen",
"Terrence",
""
],
[
"Chen",
"Chen",
""
],
[
"Wu",
"Ziyan",
""
]
] |
2410.07853 | Viktoriia Kornich | Shun Tamura, Helene M\"uller, Linus Aliani, Viktoriia Kornich | Meissner effect in non-Hermitian superconductors | null | Phys. Rev. B 111, L180503 (2025) | 10.1103/PhysRevB.111.L180503 | null | cond-mat.supr-con cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study theoretically Meissner effect in non-Hermitian systems of BCS type,
i.e., with an electron-electron interaction leading to the mean field
description in a Cooper channel, via superfluid stiffness. We show that
depending on the values of the mean fields, chemical potential, and
temperature, we obtain paramagnetic or diamagnetic Meissner effect. Notably,
positive real part of the product of mean fields guarantees diamagnetic
Meissner effect in an $s$-wave 3D non-Hermitian superconductor. Once the mean
fields are close to being anti-Hermitian, 2D $s$-, $p_x$-, and $d$-wave
superconductors exhibit interesting behaviour, including negative superfluid
stiffness giving paramagnetic Meissner effect for certain parameters. Negative
superfluid stiffness indicates instability of this collective state.
| [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 12:19:44 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Jan 2025 11:07:20 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 15:26:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Tamura",
"Shun",
""
],
[
"Müller",
"Helene",
""
],
[
"Aliani",
"Linus",
""
],
[
"Kornich",
"Viktoriia",
""
]
] |
2410.07910 | Giulia Janzen | Giulia Janzen, Juan Pablo Miranda, Jos\'e Mart\'in-Roca, Paolo
Malgaretti, Emanuele Locatelli, Chantal Valeriani, D. A. Matoz Fernandez | Active Polymer Behavior in Two Dimensions: A Comparative Analysis of
Tangential and Push-Pull Models | null | J. Chem. Phys. 162, 114905 (2025) | 10.1063/5.0243432 | null | cond-mat.soft | http://creativecommons.org/licenses/by/4.0/ | In this work, we compare the structural and dynamic behavior of active
filaments in two dimensions using tangential and push-pull models, including a
variant with passive end monomers. These models serve as valuable frameworks
for understanding self-organization in biological polymers and synthetic
materials. At low activity, all models exhibit similar behaviors. Differences
emerge in the intermediate range as activity increases, though at higher
activity levels, their behaviors converge. Importantly, adjusting for
differences in mean active force reveals nearly identical behavior across
models. Our results highlight the importance of force definitions in active
polymer simulations and provide insights into phase transitions across varying
filament configurations.
| [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 13:38:47 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Dec 2024 10:35:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Janzen",
"Giulia",
""
],
[
"Miranda",
"Juan Pablo",
""
],
[
"Martín-Roca",
"José",
""
],
[
"Malgaretti",
"Paolo",
""
],
[
"Locatelli",
"Emanuele",
""
],
[
"Valeriani",
"Chantal",
""
],
[
"Fernandez",
"D. A. Matoz",
""
]
] |
2410.08415 | Sokratis Zikas | Carolina Araujo, Daniela Paiva, Sokratis Zikas | On Gizatullin's Problem for quartic surfaces of Picard rank $2$ | final version, to appear in in Algebraic Geometry | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we determine which automorphisms of general smooth quartic
surfaces $S\subset \mathbb{P}^3$ of Picard rank $2$ are restrictions of Cremona
transformations of $\mathbb{P}^3$.
| [
{
"version": "v1",
"created": "Thu, 10 Oct 2024 23:11:33 GMT"
},
{
"version": "v2",
"created": "Wed, 6 Nov 2024 20:28:24 GMT"
},
{
"version": "v3",
"created": "Fri, 2 May 2025 20:01:50 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Araujo",
"Carolina",
""
],
[
"Paiva",
"Daniela",
""
],
[
"Zikas",
"Sokratis",
""
]
] |
2410.08535 | Javed Hussain | Javed Hussain, Saeed Ahmed, Abdul Fatah | On the Global solution and Invariance of stochastic constrained Modified
Swift-Hohenberg Equation on a Hilbert manifold | null | null | null | SIBA-001 | math.PR | http://creativecommons.org/licenses/by/4.0/ | This paper aims to investigate the stochastic generalization of the projected
deterministic constrained modified Swift-Hohenberg equation. In particular, we
prove the global well-posedness and its invariance of Hilbert submanifold i.e.
if the initial condition are chosen from submanifold then trajectories of
solutions are going to stay on manifold. The proof of global well-posedness is
based on Khashminskii test for non-explosions test for no-explosions.
Swift-Hohenberg equations belong to class of Amplitude equations that usually
describe the pattern formation in nature.
| [
{
"version": "v1",
"created": "Fri, 11 Oct 2024 05:08:58 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 08:09:44 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hussain",
"Javed",
""
],
[
"Ahmed",
"Saeed",
""
],
[
"Fatah",
"Abdul",
""
]
] |
2410.09373 | Fahimeh Habibi | Fahimeh Habibi | The Role of Outflows in dynamic of Advection Dominated Accretion Flows:
a Self-Similar Solution | null | null | null | null | astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The effects of outflow on the behavior of a viscous gaseous disc around a
compact object in an advection-dominated state are examined in this paper. We
suppose that the flow is steady, axisymmetric, and rotating. Also, we focus on
the model in which the mass, the angular momentum, and the energy can be
transported outward by outflow. Similar to the pioneering studies, we consider
a power-law function for mass inflow rate as $\dot{M} \propto r^s$. We assume
that the power index $s$ is proportional to the dimensionless thickness $H/R$
of disc. To analyze such a system, the hydrodynamic equations have extracted in
cylindrical coordinates $(r,\varphi,z)$. Then, the flow equations were
vertically integrated, and a set of self-similar solutions was got in the
radial direction. Our solutions include three essential parameters: $\lambda$,
$f$ and $\zeta$.
The influence of the outflow on the dynamics of the disc is investigated by
the $\lambda$ parameter. The degree of advection of flow is shown by the
advection parameter $f$. Also, energy extraction from the disc by the outflow
is showed by $\zeta$ parameter. Our findings demonstrate a significant
correlation between the outflow parameters, flow advection parameter, and the
temperature, thickness, and inflow-outflow rate of the disc. In addition, we
explored the influence of these parameters on the power index $s$, too. The
results of our study demonstrate that enhancing the outflow parameter or flow
advection degree increases power index $s$, while extracting more energy
through outflow decreases index $s$.
| [
{
"version": "v1",
"created": "Sat, 12 Oct 2024 05:32:43 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 11:51:36 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Habibi",
"Fahimeh",
""
]
] |
2410.09389 | Haoran Guan Mr | Haoran Guan, Yuwei Fan | Probabilistic error analysis of CholeskyQR based on columns | null | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we utilize the randomized models presented in \cite{New} and do
probabilistic error analysis of CholeskyQR2 and Shifted CholeskyQR3. We
integrate the theoretical analysis with $[\cdot]_{g}$ defined in
\cite{Columns}, providing sharper upper bounds of accuracy and better
sufficient conditions for CholeskyQR2 and Shifted CholeskyQR3 along with the
corresponding probabilities. Moreover, a probabilistic shifted item $s$ for
Shifted CholeskyQR3 is received, improving the applicability of the algorithm
while maintaining numerical stability. Numerical experiments confirm our
findings and show that such a probabilistic $s$ has good robustness in
ill-conditioned cases.
| [
{
"version": "v1",
"created": "Sat, 12 Oct 2024 06:43:48 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Feb 2025 10:26:59 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 09:14:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Guan",
"Haoran",
""
],
[
"Fan",
"Yuwei",
""
]
] |
2410.09646 | Bekir Can L\"utf\"uo\u{g}lu | M. Benarous, A. Hocine, B. C. L\"utf\"uo\u{g}lu, B. Hamil | Bounding the Wigner Deformation Parameter in Harmonically Trapped Bose
Gases | 15 pages, 2 Figures, | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | By examining the internal energy and the heat capacity of a harmonically
trapped ideal Bose gas within the Dunkl formalism, we show that the Wigner
parameter influences the slopes of these thermodynamic functions in the
critical region, reflecting its role in modifying the statistical properties of
the system. However, despite these modifications, the phase transition itself
retains the same order and critical exponents as in the standard case, in
accordance with symmetry arguments. Furthermore, upon analyzing the classical
behavior, we establish both upper and lower bounds for the Wigner parameter by
ensuring thermodynamic consistency in different temperature regimes.
| [
{
"version": "v1",
"created": "Sat, 12 Oct 2024 20:50:03 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 18:10:18 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Benarous",
"M.",
""
],
[
"Hocine",
"A.",
""
],
[
"Lütfüoğlu",
"B. C.",
""
],
[
"Hamil",
"B.",
""
]
] |
2410.09796 | Yucheng Yin | Yucheng Yin, Josh Mathews, Alvin J. K. Chua and Xian Chen | Relativistic model of binary extreme-mass-ratio inspiral systems and
their gravitational radiation | PRD published | null | 10.1103/PhysRevD.111.103007 | null | gr-qc astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A binary extreme-mass-ratio inspiral (b-EMRI) is a hierarchical triple system
consisting of a stellar-mass binary black hole (BBH) orbiting a central Kerr
supermassive black hole (SMBH). Although predicted by several astrophysical
models, b-EMRIs pose a challenge in waveform modeling due to their complex
three-body dynamics and strong relativistic effects. Here we take advantage of
the hierarchical nature of b-EMRI systems to transform the internal motion of
the small binary into global trajectories around the SMBH. This allows us to
use black hole perturbation theory to calculate both the low-frequency
gravitational waveform due to its EMRI nature and the high-frequency waveform
generated by the inner motion of the BBH. When the inner binary's separation
vanishes, our calculation recovers the standard relativistic adiabatic EMRI
waveform. Furthermore, by including the high-frequency perturbation, we find a
correction to the waveform as large as the adiabatic order when the frequency
matches the quasinormal modes (QNMs) of the SMBH, therefore supporting an
earlier proof-of-concept study claiming that the small BBH can resonantly
excite the QNMs of the SMBH. More importantly, we find that b-EMRIs can evolve
faster than regular EMRIs due to this resonant dissipation through the
high-frequency modes. These characteristics distinguish b-EMRI waveform
templates from regular EMRI templates for future space-based gravitational-wave
detectors.
| [
{
"version": "v1",
"created": "Sun, 13 Oct 2024 10:51:33 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 00:56:00 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Yin",
"Yucheng",
""
],
[
"Mathews",
"Josh",
""
],
[
"Chua",
"Alvin J. K.",
""
],
[
"Chen",
"Xian",
""
]
] |
2410.10519 | Noemi B\"uhrer | Noemi B\"uhrer, Sa\'ul Alonso-Monsalve, Matthew Franks, Till
Dieminger, Davide Sgalaberna | AI-based particle track identification in scintillating fibres read out
with imaging sensors | 23 pages, 13 figures | null | null | null | cs.LG hep-ex physics.ins-det | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents the development and application of an AI-based method for
particle track identification using scintillating fibres read out with imaging
sensors. We propose a variational autoencoder (VAE) to efficiently filter and
identify frames containing signal from the substantial data generated by SPAD
array sensors. Our VAE model, trained on purely background frames, demonstrated
a high capability to distinguish frames containing particle tracks from
background noise. The performance of the VAE-based anomaly detection was
validated with experimental data, demonstrating the method's ability to
efficiently identify relevant events with rapid processing time, suggesting a
solid prospect for deployment as a fast inference tool on hardware for
real-time anomaly detection. This work highlights the potential of combining
advanced sensor technology with machine learning techniques to enhance particle
detection and tracking.
| [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 13:59:30 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Apr 2025 09:04:10 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 07:25:31 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bührer",
"Noemi",
""
],
[
"Alonso-Monsalve",
"Saúl",
""
],
[
"Franks",
"Matthew",
""
],
[
"Dieminger",
"Till",
""
],
[
"Sgalaberna",
"Davide",
""
]
] |
2410.10807 | Youngjae Min | Youngjae Min, Navid Azizan | Hard-Constrained Neural Networks with Universal Approximation Guarantees | null | null | null | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | Incorporating prior knowledge or specifications of input-output relationships
into machine learning models has gained significant attention, as it enhances
generalization from limited data and leads to conforming outputs. However, most
existing approaches use soft constraints by penalizing violations through
regularization, which offers no guarantee of constraint satisfaction--an
essential requirement in safety-critical applications. On the other hand,
imposing hard constraints on neural networks may hinder their representational
power, adversely affecting performance. To address this, we propose HardNet, a
practical framework for constructing neural networks that inherently satisfy
hard constraints without sacrificing model capacity. Unlike approaches that
modify outputs only at inference time, HardNet enables end-to-end training with
hard constraint guarantees, leading to improved performance. To the best of our
knowledge, HardNet is the first method with an efficient forward pass to
enforce more than one input-dependent inequality constraint. It allows
unconstrained optimization of the network parameters using standard algorithms
by appending a differentiable closed-form enforcement layer to the network's
output. Furthermore, we show that HardNet retains the universal approximation
capabilities of neural networks. We demonstrate the versatility and
effectiveness of HardNet across various applications: learning with piecewise
constraints, learning optimization solvers, optimizing control policies in
safety-critical systems, and learning safe decision logic for aircraft systems.
| [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 17:59:24 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 17:58:28 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Min",
"Youngjae",
""
],
[
"Azizan",
"Navid",
""
]
] |
2410.10821 | Jingzhi Bao | Jingzhi Bao, Xueting Li, Ming-Hsuan Yang | Tex4D: Zero-shot 4D Scene Texturing with Video Diffusion Models | Project page: https://tex4d.github.io/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3D meshes are widely used in computer vision and graphics for their
efficiency in animation and minimal memory use, playing a crucial role in
movies, games, AR, and VR. However, creating temporally consistent and
realistic textures for mesh sequences remains labor-intensive for professional
artists. On the other hand, while video diffusion models excel at text-driven
video generation, they often lack 3D geometry awareness and struggle with
achieving multi-view consistent texturing for 3D meshes. In this work, we
present Tex4D, a zero-shot approach that integrates inherent 3D geometry
knowledge from mesh sequences with the expressiveness of video diffusion models
to produce multi-view and temporally consistent 4D textures. Given an
untextured mesh sequence and a text prompt as inputs, our method enhances
multi-view consistency by synchronizing the diffusion process across different
views through latent aggregation in the UV space. To ensure temporal
consistency, we leverage prior knowledge from a conditional video generation
model for texture synthesis. However, straightforwardly combining the video
diffusion model and the UV texture aggregation leads to blurry results. We
analyze the underlying causes and propose a simple yet effective modification
to the DDIM sampling process to address this issue. Additionally, we introduce
a reference latent texture to strengthen the correlation between frames during
the denoising process. To the best of our knowledge, Tex4D is the first method
specifically designed for 4D scene texturing. Extensive experiments demonstrate
its superiority in producing multi-view and multi-frame consistent videos based
on untextured mesh sequences.
| [
{
"version": "v1",
"created": "Mon, 14 Oct 2024 17:59:59 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Oct 2024 18:13:13 GMT"
},
{
"version": "v3",
"created": "Sat, 3 May 2025 12:19:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bao",
"Jingzhi",
""
],
[
"Li",
"Xueting",
""
],
[
"Yang",
"Ming-Hsuan",
""
]
] |
2410.10824 | Amartya Goswami | Amartya Goswami, Danielle Kleyn, and Kerry Porrill | On structures of the ring of arithmetical functions: prime ideals and
beyond | 12 pages. arXiv admin note: text overlap with arXiv:2302.01072 | null | null | null | math.RA | http://creativecommons.org/licenses/by/4.0/ | The aim of these notes is to study some of the structural aspects of the ring
of arithmetical functions. We prove that this ring is neither Noetherian nor
Artinian. Furthermore, we construct various types of prime ideals. We also give
an example of a semi-prime ideal that is not prime. We show that the ring of
arithmetical functions has infinite Krull dimension.
| [
{
"version": "v1",
"created": "Fri, 20 Sep 2024 21:15:28 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 16:19:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Goswami",
"Amartya",
""
],
[
"Kleyn",
"Danielle",
""
],
[
"Porrill",
"Kerry",
""
]
] |
2410.10907 | Mohammad Al-Sayed Ahmad | Mohammad Al-Sayed Ahmad, Jude Haddad | An Explainable AI Model for Predicting the Recurrence of Differentiated
Thyroid Cancer | null | null | 10.1109/JIBEC63210.2024.10932125 | null | cs.LG stat.AP | http://creativecommons.org/licenses/by/4.0/ | Thyroid carcinoma, a significant yet often controllable cancer, has seen a
rise in cases, largely due to advancements in diagnostic methods.
Differentiated thyroid cancer (DTC), which includes papillary and follicular
varieties, is typically associated with a positive prognosis in academic
circles. Nevertheless, there are still some individuals who may experience a
recurrence. This study employs machine learning, particularly deep learning
models, to predict the recurrence of DTC, with the goal of improving patient
care through personalized treatment approaches. By analysing a dataset
containing clinicopathological features of patients, the model achieved
remarkable accuracy rates of 98% during training and 96% during testing. To
improve the model's interpretability, we used techniques like LIME and Morris
Sensitivity Analysis. These methods gave us valuable insights into how the
model makes decisions. The results suggest that combining deep learning models
with interpretability techniques can be extremely useful in quickly identifying
the recurrence of thyroid cancer in patients. This can help in making informed
therapeutic choices and customizing treatment approaches for individual
patients.
| [
{
"version": "v1",
"created": "Sun, 13 Oct 2024 23:12:33 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ahmad",
"Mohammad Al-Sayed",
""
],
[
"Haddad",
"Jude",
""
]
] |
2410.11197 | Yu-Ping Wang | Machiko Hatsuda, Ond\v{r}ej Hul\'ik, William D. Linch, Warren D.
Siegel, Di Wang, Yu-Ping Wang | Strings and membranes from $\mathcal{A}$-theory five brane | 43 pages, revised version | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The $\mathcal{A}$-theory takes U-duality symmetry as a guiding principle,
with the SL(5) U-duality symmetry being described as the world-volume theory of
a 5-brane. Furthermore, by unifying the 6-dimensional world-volume Lorentz
symmetry with the SL(5) spacetime symmetry, it extends to SL(6) U-duality
symmetry. The SL(5) spacetime vielbein fields and the 5-brane world-volume
vielbein fields are mixed under the SL(6) U-duality transformation. We
demonstrate that consistent sectionings of the SL(6) $\mathcal{A}$5-brane
world-volume Lagrangian yield Lagrangians of the ${\cal T}$-string with O(D,D)
T-duality symmetry, the conventional string, the $\mathcal{M}$5-brane with
GL(4) duality symmetry, and the non-perturbative M2-brane in supergravity
theory. The GL(4) covariant Lagrangian of the $\mathcal{M}$5-brane derived in
this manner is a new, perturbatively quantizable theory.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 02:29:01 GMT"
},
{
"version": "v2",
"created": "Thu, 31 Oct 2024 16:46:53 GMT"
},
{
"version": "v3",
"created": "Mon, 25 Nov 2024 09:38:02 GMT"
},
{
"version": "v4",
"created": "Mon, 9 Dec 2024 07:08:57 GMT"
},
{
"version": "v5",
"created": "Fri, 17 Jan 2025 17:13:53 GMT"
},
{
"version": "v6",
"created": "Mon, 5 May 2025 05:06:16 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hatsuda",
"Machiko",
""
],
[
"Hulík",
"Ondřej",
""
],
[
"Linch",
"William D.",
""
],
[
"Siegel",
"Warren D.",
""
],
[
"Wang",
"Di",
""
],
[
"Wang",
"Yu-Ping",
""
]
] |
2410.11226 | Peter Eckmann | Peter Eckmann, Dongxia Wu, Germano Heinzelmann, Michael K. Gilson,
Rose Yu | MF-LAL: Drug Compound Generation Using Multi-Fidelity Latent Space
Active Learning | Accepted to ICML 2025. 9 pages, 5 figures. arXiv admin note: text
overlap with arXiv:2402.10387 | null | null | null | cs.LG q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current generative models for drug discovery primarily use molecular docking
as an oracle to guide the generation of active compounds. However, such models
are often not useful in practice because even compounds with high docking
scores do not consistently show real-world experimental activity. More accurate
methods for activity prediction exist, such as molecular dynamics based binding
free energy calculations, but they are too computationally expensive to use in
a generative model. To address this challenge, we propose Multi-Fidelity Latent
space Active Learning (MF-LAL), a generative modeling framework that integrates
a set of oracles with varying cost-accuracy tradeoffs. We train a surrogate
model for each oracle and use these surrogates to guide generation of compounds
with high predicted activity. Unlike previous approaches that separately learn
the surrogate model and generative model, MF-LAL combines the generative and
multi-fidelity surrogate models into a single framework, allowing for more
accurate activity prediction and higher quality samples. We train MF-LAL with a
novel active learning algorithm to further reduce computational cost. Our
experiments on two disease-relevant proteins show that MF-LAL produces
compounds with significantly better binding free energy scores than other
single and multi-fidelity approaches (~50% improvement in mean binding free
energy score).
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 03:15:05 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 22:21:24 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Eckmann",
"Peter",
""
],
[
"Wu",
"Dongxia",
""
],
[
"Heinzelmann",
"Germano",
""
],
[
"Gilson",
"Michael K.",
""
],
[
"Yu",
"Rose",
""
]
] |
2410.11504 | Tomasz Gradowski | Sebastian Wildowicz, Tomasz Gradowski, Paulina Figura, Igor Olczak,
Judyta Sobiech, Teodor Buchner | Physically motivated projection of the electrocardiogram -- a
feasibility study | 26 pages, 11 figures | null | 10.1016/j.bbe.2025.03.001 | null | physics.med-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present PhysECG: a physically motivated projection of the 12 lead
electrocardiogram, supported by a deep learning model trained on 21,799
recordings from the PTB-XL database and discuss its feasibility. The method
allows to evaluate the epicardial activity (inverse problem of ECG imaging)
and, in particular, to distinguish left and right ventricular activity, with
statistical spread related to localization of the septum. The observed
dyssynchrony resembles other experimental results. The foundations of the
method are based on the molecular theory of biopotentials. The heart's activity
in view of the method is decomposed into two processes: the passage of the
electric activation wavefront and the response of cardiomyocytes. We introduce
the idea of the electrode-resolved activity function, which represents the mass
of the ventricle in Phase 0 of action potential within the lead field of each
electrode. The computations are fast and robust, with excellent convergence. We
present the quality metrics for the reconstruction based on the model on the
testing set selected from the PTB database. In order to prove feasibility, we
present and discuss two healthy controls: male and female, and two pathologies:
right bundle branch block, and anterior myocardial infarction. The results
obtained using PhysECG seem to be in accordance with the changes evoked by
pathology, which has to be confirmed by subsequent clinical studies. The method
is based on ECG, and does not require reconstruction of body geometry, which
presents an affordable solution for low and middle-income countries where
access to imaging is limited.
| [
{
"version": "v1",
"created": "Tue, 15 Oct 2024 11:17:08 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Mar 2025 21:30:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wildowicz",
"Sebastian",
""
],
[
"Gradowski",
"Tomasz",
""
],
[
"Figura",
"Paulina",
""
],
[
"Olczak",
"Igor",
""
],
[
"Sobiech",
"Judyta",
""
],
[
"Buchner",
"Teodor",
""
]
] |
2410.12569 | Fabian Lenke | Fabian Lenke, Stefan Milius, Henning Urbat, Thorsten Wi{\ss}mann | Algebraic Language Theory with Effects | This is the extended version of the ICALP'25 paper | null | null | null | cs.FL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Regular languages -- the languages accepted by deterministic finite automata
-- are known to be precisely the languages recognized by finite monoids. This
characterization is the origin of algebraic language theory. In this paper, we
generalize the correspondence between automata and monoids to automata with
generic computational effects given by a monad, providing the foundations of an
effectful algebraic language theory. We show that, under suitable conditions on
the monad, a language is computable by an effectful automaton precisely when it
is recognizable by (1) an effectful monoid morphism into an effect-free finite
monoid, and (2) a monoid morphism into a monad-monoid bialgebra whose carrier
is a finitely generated algebra for the monad, the former mode of recognition
being conceptually completely new. Our prime application is a novel algebraic
approach to languages computed by probabilistic finite automata. Additionally,
we derive new algebraic characterizations for nondeterministic probabilistic
finite automata and for weighted finite automata over unrestricted semirings,
generalizing previous results on weighted algebraic recognition over
commutative rings.
| [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 13:43:37 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:28:03 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lenke",
"Fabian",
""
],
[
"Milius",
"Stefan",
""
],
[
"Urbat",
"Henning",
""
],
[
"Wißmann",
"Thorsten",
""
]
] |
2410.12689 | Antony Richard Lee Dr | Antony R. Lee, Peter Tino, Iain Bruce Styles | A distance function for stochastic matrices | 10 pages, 2 figures | null | null | null | math.PR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Motivated by information geometry, a distance function on the space of
stochastic matrices is advocated. Starting with sequences of Markov chains the
Bhattacharyya angle is advocated as the natural tool for comparing both short
and long term Markov chain runs. Bounds on the convergence of the distance and
mixing times are derived. Guided by the desire to compare different Markov
chain models, especially in the setting of healthcare processes, a new distance
function on the space of stochastic matrices is presented. It is a true
distance measure which has a closed form and is efficient to implement for
numerical evaluation. In the case of ergodic Markov chains, it is shown that
considering either the Bhattacharyya angle on Markov sequences or the new
stochastic matrix distance leads to the same distance between models.
| [
{
"version": "v1",
"created": "Wed, 16 Oct 2024 15:49:25 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 12:21:46 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lee",
"Antony R.",
""
],
[
"Tino",
"Peter",
""
],
[
"Styles",
"Iain Bruce",
""
]
] |
2410.13849 | Florian H\"ubler | Florian H\"ubler, Ilyas Fatkhullin, Niao He | From Gradient Clipping to Normalization for Heavy Tailed SGD | null | null | null | null | math.OC cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent empirical evidence indicates that many machine learning applications
involve heavy-tailed gradient noise, which challenges the standard assumptions
of bounded variance in stochastic optimization. Gradient clipping has emerged
as a popular tool to handle this heavy-tailed noise, as it achieves good
performance in this setting both theoretically and practically. However, our
current theoretical understanding of non-convex gradient clipping has three
main shortcomings. First, the theory hinges on large, increasing clipping
thresholds, which are in stark contrast to the small constant clipping
thresholds employed in practice. Second, clipping thresholds require knowledge
of problem-dependent parameters to guarantee convergence. Lastly, even with
this knowledge, current sampling complexity upper bounds for the method are
sub-optimal in nearly all parameters. To address these issues, we study
convergence of Normalized SGD (NSGD). First, we establish a parameter-free
sample complexity for NSGD of
$\mathcal{O}\left(\varepsilon^{-\frac{2p}{p-1}}\right)$ to find an
$\varepsilon$-stationary point. Furthermore, we prove tightness of this result,
by providing a matching algorithm-specific lower bound. In the setting where
all problem parameters are known, we show this complexity is improved to
$\mathcal{O}\left(\varepsilon^{-\frac{3p-2}{p-1}}\right)$, matching the
previously known lower bound for all first-order methods in all problem
dependent parameters. Finally, we establish high-probability convergence of
NSGD with a mild logarithmic dependence on the failure probability. Our work
complements the studies of gradient clipping under heavy tailed noise improving
the sample complexities of existing algorithms and offering an alternative
mechanism to achieve high probability convergence.
| [
{
"version": "v1",
"created": "Thu, 17 Oct 2024 17:59:01 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 01:23:32 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hübler",
"Florian",
""
],
[
"Fatkhullin",
"Ilyas",
""
],
[
"He",
"Niao",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.