Tao Gui

5 papers Fudan University

• BMMR: A Large-Scale Bilingual Multimodal Multi-Discipline Reasoning Dataset
• EvaLearn: Quantifying the Learning Capability and Efficiency of LLMs via Sequential Problem Solving
• INST-IT: Boosting Instance Understanding via Explicit Visual Prompt Instruction Tuning
• Pre-Trained Policy Discriminators are General Reward Models
• Understanding Parametric and Contextual Knowledge Reconciliation within Large Language Models