Yijun Tian

Reinforcement Mid-Training

ALDEN: Reinforcement Learning for Active Navigation and Evidence Gathering in Long Documents

Can visual input Be Compressed? A visual input token compression benchmark for large multimodal models

Mitigating Emergent Robustness Degradation while Scaling Graph Learning

G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering

MAPE-PPI: Towards Effective and Efficient Protein-Protein Interaction Prediction via Microenvironment-Aware Protein Embedding