experience

Research Areas: Latent Reasoning, VLA, Reinforcement Learning, Autonomous Driving

Project: ongoing research on large reasoning models and VLA.

Research Areas: LLMs, VLMs, AI Agents, Reinforcement Learning, Image Restoration

Project: developed an agentic system for complex image restoration by fine-tuning a VLM to identify degradations, employing an LLM for restoration planning, and training a reinforcement learning agent to optimize tool execution sequences (CVPR 2026).

Research Areas: LLMs, VLMs, Multimodal Alignment, Retrieval, Recommendation

Project I: proposed a quality-controllable search framework that trains an LLM for query refinement and integrates a VLM for text–image semantic matching, providing customers more aesthetic and higher-quality recommendations (ICLR 2026).

Project II: proposed the Indra Representation Hypothesis, showing that independently trained unimodal foundation models implicitly converge to a shared relational structure of reality, formulated via V-enriched Yoneda embeddings (NeurIPS 2025).

Research Areas: Graph Learning, Hashing, Domain Adaptation, Retrieval, Recommendation

Project: proposed an efficient asymmetric hashing approach for cross-modal and cross-domain search by learning an adaptive bipartite graph that models modality- and domain-level relationships, resulting in more accurate search results (TCYB 2023).