Fangzhou Lin

My research areas lie in agentic AI, multimodal AI, generative AI, especially in LLM, vision-language models, MLLM, 3D generative models, 3D perception, AI for medical, AI4Sci.


News

  • Our paper: GPS: A Probabilistic Distributional Similarity with Gumbel Priors for Set-to-Set Matching has been accepted by ICLR 2025!

  • Our paper: Loss Distillation via Gradient Matching for Point Cloud Completion with Weighted Chamfer Distance has been accepted by IEEE IROS 2024 as an Oral presentation!

  • Our paper: Understanding Hyperbolic Metric Learning through Hard Negative Sampling has been accepted by IEEE WACV 2024!

  • Our paper: InfoCD: A Contrastive Chamfer Distance Loss for Point Cloud Completion has been accepted by NeurIPS 2023!

  • Our paper: Hyperbolic Chamfer Distance for Point Cloud Completion has been accepted by IEEE ICCV 2023!


Service

Reviewer of ICLR, NeurIPS, CVPR, ICCV, IROS, MICCAI

Home / Publications / CV