Cover Image for Three Frontiers of Scalable RL for LLMs
Cover Image for Three Frontiers of Scalable RL for LLMs
Hosted By
1 Went

Three Frontiers of Scalable RL for LLMs

Hosted by NICE AI Talk
YouTube
Registration
Past Event
Welcome! To join the event, please register below.
About Event

​NICE Talk 172 invites Bingxiang He @HBX_hbx, PhD student at Tsinghua University @TsinghuaNLP, to share Three Frontiers of Scalable RL for LLMs.

​Talk Time ⏰ EST: 5.15 22:00~23:00

​🤠 Can RL advance model capabilities without any supervised signals?

​🧐 Three Frontiers, One Map: Charting the Feasible Region of Scalable RL Matters More Than Inventing Another Trick

​😈 Explicit length penalties and more lenient verifiers both led to significant performance degradation.

​😈 Switching to a higher-scoring teacher can paradoxically shrink—or even reverse—student gains.

​Work will related in the talk:

​🌟JustRL: thunlp/JustRL

​🌟Unsupervised RLVR: https://github.com/PRIME-RL/TTRL/tree/urlvr-dev

​🌟Rethinking On-Policy Distillation: thunlp/OPD

​Host: Cheng Qian, PhD at UIUC

​#AI #LLM #scalinglaw #model

Hosted By
1 Went