Stefan Schweter PRO

stefan-it's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a collection about 18 hours ago

BioClinical ModernBERT

Collection

3 items • Updated about 19 hours ago • 7

upvoted 3 papers 1 day ago

BioClinical ModernBERT: A State-of-the-Art Long-Context Encoder for Biomedical and Clinical NLP

Paper • 2506.10896 • Published 1 day ago • 1

Domain2Vec: Vectorizing Datasets to Find the Optimal Data Mixture without Training

Paper • 2506.10952 • Published 1 day ago • 20

Magistral

Paper • 2506.10910 • Published 1 day ago • 26

upvoted a collection 2 days ago

Institutional Books

Collection

A growing corpus of public domain books from library collections, seeded by Harvard Library. • 3 items • Updated 2 days ago • 1

upvoted a paper 3 days ago

Institutional Books 1.0: A 242B token dataset from Harvard Library's collections, refined for accuracy and usability

Paper • 2506.08300 • Published 4 days ago • 6

upvoted 2 papers 8 days ago

The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text

Paper • 2506.05209 • Published 9 days ago • 36

Static Word Embeddings for Sentence Semantic Representation

Paper • 2506.04624 • Published 9 days ago • 3

upvoted 2 papers 10 days ago

Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Training

Paper • 2506.01732 • Published 12 days ago • 2

XToM: Exploring the Multilingual Theory of Mind for Large Language Models

Paper • 2506.02461 • Published 11 days ago • 1

upvoted an article 11 days ago

view article

Article

No GPU left behind: Unlocking Efficiency with Co-located vLLM in TRL

By and 5 others •

11 days ago

• 49

upvoted 3 papers 11 days ago

EmoBench-UA: A Benchmark Dataset for Emotion Detection in Ukrainian

Paper • 2505.23297 • Published 16 days ago • 1

LLM in the Loop: Creating the PARADEHATE Dataset for Hate Speech Detoxification

Paper • 2506.01484 • Published 12 days ago • 5

Novel Benchmark for NER in the Wastewater and Stormwater Domain

Paper • 2506.01938 • Published 12 days ago • 1

upvoted a collection 12 days ago

Common Pile v0.1

Collection

All resources related to Common Pile v0.1, an 8TB dataset of public domain and openly licensed text • 4 items • Updated 8 days ago • 25

upvoted a collection 18 days ago

ModernGBERT

Collection

3 items • Updated 18 days ago • 4

upvoted a paper 18 days ago

ModernGBERT: German-only 1B Encoder Model Trained from Scratch

Paper • 2505.13136 • Published 26 days ago • 21

upvoted 3 papers 19 days ago

On Relation-Specific Neurons in Large Language Models

Paper • 2502.17355 • Published Feb 24 • 9

Understanding Gated Neurons in Transformers from Their Input-Output Functionality

Paper • 2505.17936 • Published 22 days ago • 1

Language Mixing in Reasoning Language Models: Patterns, Impact, and Internal Causes

Paper • 2505.14815 • Published 25 days ago • 1