6 papers
MotionHalluc: Diagnosing Kinematic Hallucinations in Fine-Grained Motion Reasoning
Weile Guo, Shenghong He, Danying Mo +3
Motion instruction generation in cross-video comparison aims to produce corrective feedback that describes the differences between a query and a reference motion. However, existing…
From Verdict to Process: Agentic Reinforcement Learning for Multi-Stage Fact Verification
Rongxin Yang, Shenghong He, Siyuan Zhu +1
Recent approaches combining Large Language Models (LLMs) with retrieval-augmented reasoning have shown promise for automated fact verification. To process complex claims, these ver…
Adaptive Coarse-to-Fine Subgoal Refinement for Long-Horizon Offline Goal-Conditioned Reinforcement Learning
Kaiqiang Ke, Shenghong He, Chengdong Xu +3
Offline goal-conditioned reinforcement learning (GCRL) is challenging in long-horizon tasks, where distant state--goal pairs provide weak supervision and value estimates become vul…
Advantage-based Temporal Attack in Reinforcement Learning
Shenghong He
Extensive research demonstrates that Deep Reinforcement Learning (DRL) models are susceptible to adversarially constructed inputs (i.e., adversarial examples), which can mislead th…
Gradient Inversion in Federated Reinforcement Learning
Shenghong He
Federated reinforcement learning (FRL) enables distributed learning of optimal policies while preserving local data privacy through gradient sharing.However, FRL faces the risk of…
Model-Based Offline Reinforcement Learning with Reliability-Guaranteed Sequence Modeling
Shenghong He
Model-based offline reinforcement learning (MORL) aims to learn a policy by exploiting a dynamics model derived from an existing dataset. Applying conservative quantification to th…