Zhejiang University
College of Computer Science and Technology · Pervasive Computing Lab
I am a Ph.D. student in Computer Science and Technology at Zhejiang University. My research interests include omni-modal large models and temporal foundation models. Before joining Zhejiang University, I worked on multimodal large language models for intelligent cockpits at NIO and received my M.Eng. from Nanjing University of Science and Technology.
Open to research collaboration, academic discussion, and career opportunities.
College of Computer Science and Technology · Pervasive Computing Lab
PCALab · Research in multimodal learning, natural language processing, and data mining.
Led core development of an on-device omni-modal model and a streaming video interaction system for intelligent cockpits. Built end-to-end multimodal training and evaluation pipelines, with a focus on efficient perception and proactive, event-driven interaction.
Researched open-world semi-supervised class discovery and supported job data processing and matching. The proposed method was accepted by ICDM 2023.
Introduces a multi-granularity distillation approach to mitigate modality imbalance while preserving instance-level semantics and structure for cross-modal and unimodal retrieval.
Proposes SSOC, a semi-supervised method for discovering novel classes in open-world settings by explicitly modeling class centers and aligning prediction and embedding spaces.
Contributed to an open-source toolkit built with PaddlePaddle and PyTorch for joint and cross-modal learning, including retrieval, classification, and image captioning.
GitHub Repository ↗Multimodal resume analysis application developed with Baidu’s Talent Intelligence Center.