5 papers
Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL
Tong Zheng, Skylar Zhai, Zhan Cheng +7
Multi-reward reinforcement learning trains large language models to satisfy multiple behavioral objectives simultaneously. Reward-wise normalization, as used in GDPO, preserves rew…
RolloutFaith: Auditing Persistent Internal Interventions in Visual World Model
Junchi Yao, Ziyi Wang, Youling Huang +1
Interpretability methods such as probes, activation patches and learned editors are designed to reveal or modify a model's current computation. World models pose a harder requireme…
TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search
Zhuofan Shi, Mingzhe Ma, Lu Wang +8
Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis. A central challenge is deciding how to search w…
Beyond State Consistency: Behavior Consistency in Text-Based World Models
Youling Huang, Guanqiao Chen, Junchi Yao +8
World models have been emerging as critical components for assessing the consequences of actions generated by interactive agents in online planning and offline evaluation. In text-…
Cross-Domain Policy Optimization via Bellman Consistency and Hybrid Critics
Ming-Hong Chen, Kuan-Chen Pan, You-De Huang +2
Cross-domain reinforcement learning (CDRL) is meant to improve the data efficiency of RL by leveraging the data samples collected from a source domain to facilitate the learning in…