Zongyu Lin

Learning to Rank Chain-of-Thought: Using a Small Model

LLMs as Scalable, General-Purpose Simulators For Evolving Digital Agent Training

Learning Diffusion Models with Flexible Representation Guidance

VideoPhy: Evaluating Physical Commonsense for Video Generation

MMEgo: Towards Building Egocentric Multimodal LLMs for Video QA

Contradiction Retrieval via Contrastive Learning with Sparsity

Contradiction Retrieval Via Sparse-Aware Sentence Embedding

QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search

Q* Agent: Optimizing Language Agents with Q-Guided Exploration