Paper published: 'Feature Transformation by Semi-AR and reward-guided diffusion' accepted to NeurIPS 2025.
Paper published: 'MMTok: Multimodal Coverage Maximization for Efficient VLM Inference' launched on arXiv.
Paper published: 'LiveMCP-101: a new benchmark testing AI agents’ real-world tool-use' released on arXiv.
Paper published: 'LogicIF: Complex Logical Instruction Generation' released on arXiv.
Comprehensive blog post published: 'TimesCLIP: our multimodal approach to time series forecasting with CLIP'.
Paper published: 'MLLM-Tool' accepted to WACV 2024.
Paper published: 'WeakSVR' accepted to CVPR 2023.
Paper published: 'TransRAC' accepted as oral presentation to CVPR 2022.
Education
Master's Degree: ShanghaiTech University; Advisor: Professor Shenghua Gao
Background
Research Interests: Multimodal Learning, VLM, LLM Agent; Professional Field: Computer Vision, Natural Language Processing, and Machine Learning; Brief Introduction: Currently an independent researcher.