News
Published:
18/05/2025: Our work on “On-Policy RL Meets Off-Policy Experts in Fine-tuning LLM” was posted on ArXiv.
19/05/2025: Our work on latent computation in LLM was posted on ArXiv.
18/04/2025: We release a new framework for reinforce fine-tuning of LLM at modelscope/Trinity-RFT and ArXiv. Welcome for any discussion!
12/04/2025: Our paper on FL with selective layer fine-tuning was accepted to ISIT 2025! The paper and code are public now.