Gang Chen

On the Fragility of Graph Backdoor Defenses: A Robust Strategy via Layer-wise Feature Divergence

Merge-of-Thought Distillation

TraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning

Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecasting

Table as a Modality for Large Language Models

In-Context Adaptation to Concept Drift for Learned Database Operations

Bridging the Semantic Gap Between Text and Table: A Case Study on NL2SQL

Injecting Learnable Table Features into LLMs

Learned Data Transformation: A Data-centric Plugin for Enhancing Time Series Forecasting

FedSR: Frequency-Aware Enhancement for Diffusion-based Image Super-Resolution