5 4 19

hassenhamdi

AI & ML interests

None yet

Recent Activity

new activity 1 day ago

ZennyKenny/tactical-military-reasoning-v.1.0:🚩 Report: Illegal or restricted content

new activity 1 day ago

ZennyKenny/tactical-military-reasoning-v.1.0:🚩 Report: Illegal or restricted content

replied to their post 1 day ago

I want to remind community that, making tech safe and beneficial and using it responsibly is a collective responsibility and I urge every member of community that value rational and ethical use of tech, AI etc. to take action, stand and speak against any conduct that might cause things to shift toward undesired path, Yesterday a dataset been uploaded on huggingface ,your favorite daily place for AI dataset, model, post ,research, learning about AI etc, for military applications which is against ethics and responsible AI, huggingface policies as stated in their restricted content list: ``` 1. Unlawful or illegal Content ... Content promoting high-risk illegal activities (weapons development, illegal substances, scams, gambling, pseudo-pharmaceuticals, plagiarism, etc.). ... 3. Harmful or Abusive Content Terrorist Content or Content that glorifies **violence**, suffering, or humiliation. ... we may also moderate other types of Content in response to evolving challenges posed by advancements in Machine Learning. ``` I urge community to report any content against ethical use of tech stack including AI , keeping huggingface good place for AI and data enthusiasts. Here the link to the dataset: https://huggingface.co/datasets/ZennyKenny/tactical-military-reasoning-v.1.0 Take action report it, let the platform stay an enjoyable place for whatever you are using it for AI , Data etc.

View all activity

Organizations

hassenhamdi's activity

replied to their post 1 day ago

@JLouisBiz I have read your comment but it does not make any sense , what is the purpose of regulations and law if there is no limitations to what one can do.

We are not talking about chair here or any ordinary object for innocent usage, we are talking about war tech developement, are a chair and war tools that made primary for destruction and killing the same !!!?

They are not , probably if compared it with a gun it might be more comparable , you can have a gun to protect yourself but you need a permission for possessing a gun as us citizen or you get arrested for possessing unathorized item, you ask why , to unsure public safety, but even with such procedure there still horrifying incident that happens, we are not talking about something similar to chair here, the analogy is poorly representitive of the present situation.

And it is not about opensource and free software it is about not developing war tools.

Take an example someone using their computer for piracy , cyber threats , scams fraudelent activities etcs., do you let them just go their way or some actions need be made to protect people , war tech are far worse than any thing mentioned earlier.

posted an update 2 days ago

Post

331

1. Unlawful or illegal Content
...
Content promoting high-risk illegal activities (weapons development, illegal substances, scams, gambling, pseudo-pharmaceuticals, plagiarism, etc.).
...
3. Harmful or Abusive Content
Terrorist Content or Content that glorifies **violence**, suffering, or humiliation.
...
we may also moderate other types of Content in response to evolving challenges posed by advancements in Machine Learning.

I urge community to report any content against ethical use of tech stack including AI , keeping huggingface good place for AI and data enthusiasts.
Here the link to the dataset: ZennyKenny/tactical-military-reasoning-v.1.0
Take action report it, let the platform stay an enjoyable place for whatever you are using it for AI , Data etc.

3 replies

posted an update 3 days ago

Post

364

It seems that huggingface consider deepfakes content a violation/ misconduct but not encouraging harmful activity or application such using AI for developing war technology.
How shameful.

replied to Dcas89's post 3 days ago

It will be nice addition to make paper that include benchmarking against different task and models.

replied to ZennyKenny's post 3 days ago

It is about morality , what you made could be not of a big significance for military research and applications but may Allah forbid noramlize the idea of use AI and IT or any other thing in general to things such deceptive content , military and etc.
Such things are likely to be misused so such things should not be passed as 'OK' but be opposed to collectively.
I am reporting this.

replied to ZennyKenny's post 3 days ago

We are not in need for applying AI to military to be used for war to kill civilians (women ,children )and innocents people.
Are you aware of what's going on in Palestine due to misuse of AI to target civilans ???

reacted to sr-rai's post with 🤗 22 days ago

Post

2659

ExLlamaV3 is out. And it introduces EXL3 - a new SOTA quantization format!

"The conversion process is designed to be simple and efficient and requires only an input model (in HF format) and a target bitrate. By computing Hessians on the fly and thanks to a fused Viterbi kernel, the quantizer can convert a model in a single step, taking a couple of minutes for smaller models, up to a few hours for larger ones (70B+) (on a single RTX 4090 or equivalent GPU.)"

Repo: https://github.com/turboderp-org/exllamav3

1 reply

replied to lianghsun's post 24 days ago

it will be great addition if it does maybe with an editor like reddit 🤖 post .

replied to lianghsun's post 24 days ago

fix the link to github it has ')**' at the end.
Also it appears that huggingface posts don't handle markdowns.

reacted to MohamedRashad's post with 🤝 29 days ago

Post

2572

I collected the recitations of the holy quran from 20 different reciters and uploaded the full dataset here:
MohamedRashad/Quran-Recitations

Check it out 🥷

reacted to singhsidhukuldeep's post with 🧠 2 months ago

Post

3512

O1 Embedder: Transforming Retrieval Models with Reasoning Capabilities

Researchers from University of Science and Technology of China and Beijing Academy of Artificial Intelligence have developed a novel retrieval model that mimics the slow-thinking capabilities of reasoning-focused LLMs like OpenAI's O1 and DeepSeek's R1.

Unlike traditional embedding models that directly match queries with documents, O1 Embedder first generates thoughtful reflections about the query before performing retrieval. This two-step process significantly improves performance on complex retrieval tasks, especially those requiring intensive reasoning or zero-shot generalization to new domains.

The technical implementation is fascinating:

- The model integrates two essential functions: Thinking and Embedding
- It uses an "Exploration-Refinement" data synthesis workflow where initial thoughts are generated by an LLM and refined by a retrieval committee
- A multi-task training method fine-tunes a pre-trained LLM to generate retrieval thoughts via behavior cloning while simultaneously learning embedding capabilities through contrastive learning
- Memory-efficient joint training enables both tasks to share encoding results, dramatically increasing batch size

The results are impressive - O1 Embedder outperforms existing methods across 12 datasets in both in-domain and out-of-domain scenarios. For example, it achieves a 3.9% improvement on Natural Questions and a 3.0% boost on HotPotQA compared to models without thinking capabilities.

This approach represents a significant paradigm shift in retrieval technology, bridging the gap between traditional dense retrieval and the reasoning capabilities of large language models.

What do you think about this approach? Could "thinking before retrieval" transform how we build search systems?

replied to wassemgtk's post 2 months ago

Would like to see performance on well known benchmark datasets.

reacted to wassemgtk's post with 🧠 2 months ago

Post

1880

# GESAL: Real-Time Adaptation for LLMs

We’re excited to unveil **Graph-Enhanced Singular Adaptive Learning (GESAL)**, a framework that lets LLMs like meta-llama/Llama-3.2-1B adapt in real time using user feedback. Check out the code and white paper on GitHub!

🔗 **Code**: [https://github.com/writer/AI-Adaptive-Learning-GESAL](https://github.com/writer/AI-Adaptive-Learning-GESAL)

---

## Why GESAL?

Static LLMs struggle to adapt without heavy retraining. GESAL solves this with:
- **SVF**: Adapts weights via \( W' = U (\Sigma \cdot z) V^T \), using few parameters.
- **Graph Memory**: Stores adaptations in nodes for scalability.
- **RL**: Updates via \( J(z) = \mathbb{E}[\log \pi_z(y|x) r] \) based on feedback.

---

## How It Works

Ask "How many R’s in ‘strawberry’?" If it says "2" and you say "no," GESAL learns to say "3" next time, avoiding repeats.

---

## Try It

Built with Hugging Face’s transformers:

pip install transformers torch numpy
python Adaptive_Learning_(GESAL).py

Needs a Hugging Face token for Llama-3.2-1B.

---

## Results

GESAL hits 95% accuracy after 5 feedbacks vs. LoRA’s 70%. It’s efficient (~0.5M params) and scalable.

15 replies

reacted to freddyaboulton's post with 🚀 2 months ago

Post

3264

Getting WebRTC and Websockets right in python is very tricky. If you've tried to wrap an LLM in a real-time audio layer then you know what I'm talking about.

That's where FastRTC comes in! It makes WebRTC and Websocket streams super easy with minimal code and overhead.

Check out our org: hf.co/fastrtc

reacted to KonradSzafer's post with 👀 2 months ago

Post

1877

I’ve been experimenting with a “Tech Tree” to make ML research more systematic and transparent—turns out it helped me spot hidden interactions between experiments and share progress more easily. I wrote a short blog post with examples and insights! KonradSzafer/tech_tree_blog

reacted to MohamedRashad's post with 👍 3 months ago

Post

3322

Today is a big day for the Arabic Language,

We have Navid-AI/The-Arabic-Rag-Leaderboard,
an Update for OALL/Open-Arabic-LLM-Leaderboard
and the release of atlasia/darija-chatbot-arena

All of this announcements was under 12 hours of time 🤯

replied to Abhaykoul's post 4 months ago

I like the website UI , looks neat. experience might become better if removing the loading , and the example link to example instead of 404 error page. Overall , I hope that people use your project reasonably. @Abhaykoul

replied to Abhaykoul's post 4 months ago

AI and emotion , honestly I genuinely can't comprehend that, the world contains 8B human being (some of them are demons in the flesh of humans), and seeking to make machine more emotional, whereas in the real world people tend to act like machines, I think there real need for people is to be connected to Allah (God) then they won't have any emptiness nor the need for any other emotional support, faith (is both emotional and rational) is a primary instinctive need for human being, I think people should stem the grain of faith in their heart to find the pleasure of it and the true meaning of love. Genuinely I advice people to know who is Allah and what does it mean to believe.

reacted to TuringsSolutions's post with 👍 7 months ago

Post

3199

I solved the biggest math problem associated with the Attention Mechanism. it works, better than I ever expected. Test it all yourself. Everything you need is linked from this video: https://youtu.be/41dF0yoz0qo

Sorry the audio quality sucks, I will buy a new microphone today. Why does some moron like me solve these things and not you? I know more about how computers work than you do, that's it. Swarm algorithms were big in the 90's and early 2000's. Computers were absolute dog doo doo then in one specific way, compared to now. That one way, which everyone overlooks, is the entire secret behind why swarm algorithms are so good.