BigScience Workshop

non-profit

https://bigscience.huggingface.co

bigscience-workshop

AI & ML interests

A one-year long research workshop on large language models: the Summer of Language Models 21 🌸

Recent Activity

odegiber authored a paper about 1 month ago

Scaling Low-Resource MT via Synthetic Data Generation with LLMs

odegiber authored a paper about 1 month ago

Open Machine Translation for Esperanto

israel authored a paper about 1 month ago

CaMMT: Benchmarking Culturally Aware Multimodal Machine Translation

View all activity

authored 2 papers about 1 month ago

Scaling Low-Resource MT via Synthetic Data Generation with LLMs

Paper • 2505.14423 • Published May 20, 2025 • 2

Open Machine Translation for Esperanto

Paper • 2603.29345 • Published Mar 31

authored a paper about 2 months ago

Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning

Paper • 2601.18722 • Published Jan 26

in bigscience/bloom 2 months ago

[SPAM] Deleted

#289 opened 2 months ago by

authored a paper 2 months ago

RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation

Paper • 2603.09723 • Published Mar 10 • 7

authored a paper 2 months ago

LLM2Vec-Gen: Generative Embeddings from Large Language Models

Paper • 2603.10913 • Published Mar 11 • 44

stas

posted an update 2 months ago

Post

235

Good news! Ulysses Sequence Parallelism from the Snowflake AI Research and the Deepspeed teams has been integrated into
HuggingFace Trainer, Accelerate and TRL

For extensive details please see this writeup:
https://huggingface.co/blog/ulysses-sp

Thanks a lot to Kashif Rasul for helping make it happen. Also the others in the HF team who helped with integration.

in bigscience/bloom 2 months ago

pretokenizer Regex issues?

#278 opened almost 2 years ago by

in bigscience/bloom 3 months ago

Test PR

#286 opened 3 months ago by

Test discussion

#287 opened 3 months ago by

Test discussion

#288 opened 3 months ago by

authored a paper 3 months ago

References Improve LLM Alignment in Non-Verifiable Domains

Paper • 2602.16802 • Published Feb 18 • 2

submitted 2 papers to Daily Papers 3 months ago

Single-minus gluon tree amplitudes are nonzero

Paper • 2602.12176 • Published Feb 12 • 8

Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL

Paper • 2602.03773 • Published Feb 3 • 13

authored a paper 4 months ago

PingPong: A Natural Benchmark for Multi-Turn Code-Switching Dialogues

Paper • 2601.17277 • Published Jan 24 • 6

stas

authored a paper 4 months ago

Arctic Long Sequence Training: Scalable And Efficient Training For Multi-Million Token Sequences

Paper • 2506.13996 • Published Jun 16, 2025 • 1

authored a paper 4 months ago

Patient-Similarity Cohort Reasoning in Clinical Text-to-SQL

Paper • 2601.09876 • Published Jan 14 • 7

authored a paper 5 months ago

TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior

Paper • 2512.20757 • Published Dec 23, 2025 • 18

in bigscience/bloomz-560m 5 months ago

Fails to load with transformers v4.57+

#14 opened 5 months ago by

authored a paper 5 months ago

Economies of Open Intelligence: Tracing Power & Participation in the Model Ecosystem

Paper • 2512.03073 • Published Nov 27, 2025 • 7