8 papers
Contextual Credit Policy Optimization
Huiwen Han, Ziyao Zeng, Zihao Zhu +6
Group-based reinforcement learning enables large language model agents to learn from sparse outcome rewards in long-horizon tasks. Recent methods derive step-level credit by compar…
Disentangling Task-Relevant Latent Dynamics for Reward Modeling
Dayou Li, Jionghao Wang, Peihao Wang +9
Reward design remains a major bottleneck for reinforcement learning in robotic manipulation. Handcrafted dense rewards can compensate for the limited guidance provided by sparse re…
Explicit Trajectory Diversity for RL-Based Post-Training of LLM Agents
Huaiyu Fu, Heng Cao, Hao Wang +2
LLM agents often admit multiple high-quality solutions to the same task, differing in reasoning structure, tool-use pattern, or interaction trajectory. Yet existing notions of dive…
Boosting Deepresearch and LongContext Ability with Self-Generated Deepresearch Rollouts Traces
Zihan Wang, Hao Wang, Boyuan Jiang +8
Deepresearch (DR) agents interact with real-world web environments through multi-turn search and visit, causing their contexts to grow rapidly over time. We observe that, even afte…
Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering
Hao Wang, Jiuzhou Lei, Dayou Li +5
Behavior-cloned policies often learn multiple behavior modes from demonstration datasets, including modes that are unsafe or otherwise undesired at deployment. For example, a polic…
Learning Actionable Manipulation Recovery via Counterfactual Failure Synthesis
Dayou Li, Jiuzhou Lei, Hao Wang +6
While recent foundation models have significantly advanced robotic manipulation, these systems still struggle to autonomously recover from execution errors. Current failure-learnin…
DEEPMED: Building a Medical DeepResearch Agent via Multi-hop Med-Search Data and Turn-Controlled Agentic Training & Inference
Zihan Wang, Hao Wang, Shi Feng +6
Medical reasoning models remain constrained by parametric knowledge and are thus susceptible to forgetting and hallucinations. DeepResearch (DR) models ground outputs in verifiable…
GTPO and GRPO-S: Token and Sequence-Level Reward Shaping with Policy Entropy
Hongze Tan, Zihan Wang, Jianfei Pan +7
Reinforcement Learning (RL) is pivotal for enhancing Large Language Model (LLM) reasoning, yet mainstream algorithms such as GRPO and DAPO remain constrained by a coarse-grained cr…