DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models Paper • 2603.26164 • Published Mar 27 • 362
Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model Paper • 2603.21986 • Published Mar 23 • 124
view article Article Welcome Gemma 4: Frontier multimodal intelligence on device +5 26 days ago • 880
Gemma 4 Collection Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. • 28 items • Updated 5 days ago • 161
FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization Paper • 2603.19835 • Published Mar 20 • 348
view article Article TRL v1.0: Post-Training Library Built to Move with the Field +2 28 days ago • 50