Stefan Schweter's picture

Stefan Schweter PRO

stefan-it

·

AI & ML interests

Flair Library 💕, NER & PoS Tagging, LM Pretraining (mostly encoder-only & encoder-decoder), Historical Language Models, German Language Models

Recent Activity

upvoted a collection about 4 hours ago

Eynollah models

upvoted a paper about 4 hours ago

Large Means Left: Political Bias in Large Language Models Increases with Their Number of Parameters

commented on a paper about 4 hours ago

Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs

View all activity

Organizations

stefan-it's activity

upvoted a collection about 4 hours ago

Eynollah models

Eynollah models for document image processing and layout analysis tasks. • 14 items • Updated Mar 27 • 3

upvoted 2 papers about 4 hours ago

Large Means Left: Political Bias in Large Language Models Increases with Their Number of Parameters

Paper • 2505.04393 • Published about 23 hours ago • 1

Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs

Paper • 2505.04519 • Published about 20 hours ago • 1

upvoted a paper 2 days ago

ReplaceMe: Network Simplification via Layer Pruning and Linear Transformations

Paper • 2505.02819 • Published 3 days ago • 23

upvoted a paper 6 days ago

Aleph-Alpha-GermanWeb: Improving German-language LLM pre-training with model-based data curation and synthetic data generation

Paper • 2505.00022 • Published 14 days ago • 1

upvoted a paper 9 days ago

Efficient Domain-adaptive Continual Pretraining for the Process Industry in the German Language

Paper • 2504.19856 • Published 10 days ago • 1

upvoted a paper 10 days ago

Even Small Reasoners Should Quote Their Sources: Introducing the Pleias-RAG Model Family

Paper • 2504.18225 • Published 13 days ago • 12

upvoted a paper 14 days ago

A Post-trainer's Guide to Multilingual Training Data: Uncovering Cross-lingual Transfer Dynamics

Paper • 2504.16677 • Published 15 days ago • 1

upvoted a paper 24 days ago

ModernBERT or DeBERTaV3? Examining Architecture and Data Influence on Transformer Encoder Models Performance

Paper • 2504.08716 • Published 27 days ago • 10

upvoted a paper 28 days ago

OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens

Paper • 2504.07096 • Published 29 days ago • 73

upvoted a paper 29 days ago

Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation

Paper • 2504.06225 • Published 30 days ago • 1

upvoted 4 papers about 1 month ago

Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

Paper • 2504.03624 • Published Apr 4 • 13

Boundless Byte Pair Encoding: Breaking the Pre-tokenization Barrier

Paper • 2504.00178 • Published Mar 31 • 1

Multimodal LLMs for OCR, OCR Post-Correction, and Named Entity Recognition in Historical Documents

Paper • 2504.00414 • Published Apr 1 • 1

Overcoming Vocabulary Constraints with Pixel-level Fallback

Paper • 2504.02122 • Published Apr 2 • 2

upvoted a collection about 1 month ago

E3C-Projected

This collection contains the projected datasets of English layer one of e3c into Greek, Italian, Polish, Slovak, and Slovenian • 11 items • Updated Jan 8 • 1

upvoted a paper about 1 month ago

State Fourier Diffusion Language Model (SFDLM): A Scalable, Novel Iterative Approach to Language Modeling

Paper • 2503.17382 • Published Mar 16 • 1

upvoted a paper about 2 months ago

BLOOM: A 176B-Parameter Open-Access Multilingual Language Model

Paper • 2211.05100 • Published Nov 9, 2022 • 31

upvoted an article about 2 months ago

Article

Xet is on the Hub

Mar 18

• 47

upvoted a paper about 2 months ago

xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference

Paper • 2503.13427 • Published Mar 17 • 3