7 papers
Temporal GRPO: Beyond Trajectory-Level Credit in Vision-Language-Action Reinforcement Learning
Yao Zhou, Hang Gao, Fengge Wu +2
Outcome-driven reinforcement learning offers a scalable way to post-train vision-language-action (VLA) policies from sparse task-success feedback. In common GRPO-based VLA post-tra…
LLM-Enhanced Hierarchical Heterogeneous Graph Representation Learning for Malicious Python Package Detection
Hang Gao, Xiaoyu Chen, Baoquan Cui +4
Malicious Python packages have become a major threat to software supply chain ecosystems due to the widespread adoption of open-source repositories such as PyPI. Existing learning-…
Efficient Code Analysis via Graph Representation Learning-Guided Large Language Models
Hang Gao, Tao Peng, Baoquan Cui +4
Large Language Models (LLMs) have significantly advanced code analysis tasks, yet they struggle to detect malicious behaviors fragmented across files, whose intricate dependencies…
Causal Front-Door Adjustment for Robust Jailbreak Attacks on LLMs
Yao Zhou, Zeen Song, Wenwen Qiang +4
Safety alignment mechanisms in Large Language Models (LLMs) often operate as latent internal states, obscuring the model's inherent capabilities. Building on this observation, we m…
LLM Enhancers for GNNs: An Analysis from the Perspective of Causal Mechanism Identification
Hang Gao, Wenxuan Huang, Fengge Wu +3
The use of large language models (LLMs) as feature enhancers to optimize node representations, which are then used as inputs for graph neural networks (GNNs), has shown significant…
Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning
Hang Gao, Chenhao Zhang, Tie Wang +4
Large Language Models (LLMs) have achieved remarkable success across various domains. However, they still face significant challenges, including high computational costs for traini…