Published 30+ papers in top-tier venues including CVPR, ICML, SIGIR, ECCV, IJCAI, AAAI, ACL, and NeurIPS.
Three papers accepted to ICCV 2025.
Served as project lead or corresponding author on key projects: Ming-Omni, M2-omni, DynFocus, SegAgent, M2-Encoder, M2-RAAP, SyCoCa, etc.
Holds multiple U.S. and Chinese patents, including:
- U.S. Patent: 'Method and system for facilitating recognition of vehicle parts based on a neural network' (US 11,475,660)
- U.S. Patent Application: 'Coding apparatuses, and data processing methods and apparatuses' (2024)
- Over ten Chinese invention patents covering multimodal model training, video understanding, image processing, emotion tracking, and implicit authentication.