ARISE: Adapting to Evolving Capability Gaps in Agentic Reinforcement Learning
Adaptive Rubric-Skill Co-Evolution (ARISE), a reinforcement learning framework that uses rollout evidence to continually adapt evaluation criteria, exploration guidance, and training priorities, and demonstrates that ARISE successfully enhances both overall task performance and training efficiency.
Kun Feng, Yu-Chen Fang, Yih-Yang Tan et al.
· 0 citations