Jing Tang

Assistant Professor@The Hong Kong University of Science and Technology (Guangzhou)·中国·OpenReview

研究方向

Large language models · Machine learning · bandits · Blockchains · submodular optimization · Big data management and analytics · Social network and graph analysis

Jing Tang

Reinforcing Diffusion Models by Direct Group Preference Optimization

Exposing Weaknesses of Large Reasoning Models through Graph Algorithm Problems

Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning

Mitigating Coordinate Prediction Bias from Positional Encoding Failures

Are LLMs Really Not Knowledgeable? Mining the Submerged Knowledge in LLMs' Memory

IUT-Plug: A Plug-in tool for Interleaved Image-Text Generation

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration

Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers

TreeRPO: Tree Relative Policy Optimization

Reward-Instruct: A Reward-Centric Approach to Fast Photo-Realistic Image Generation

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

KaSA: Knowledge-Aware Singular-Value Adaptation of Large Language Models

Decoupled Graph Energy-based Model for Node Out-of-Distribution Detection on Heterophilic Graphs

Improving LLMs‘ Generalized Reasoning Abilities by Graph Problems

Noise Consistency Training: A Native Approach for One-step Generator in Learning Additional Controls

You Only Sample Once: Taming One-Step Text-to-Image Synthesis by Self-Cooperative Diffusion GANs

Process-Driven Autoformalization in Lean 4