Assistant Professor, Zhejiang University
2 papers at NeurIPS 2025
We propose RepBlend, a multimodal dataset distillation framework that mitigates modality collapse via representation blending and symmetric projection matching, achieving superior performance and generalization.
This paper presents 3D-RAD, a large-scale dataset designed to advance 3D Med-VQA using radiology CT scans. The 3D-RAD dataset encompasses six diverse VQA tasks.
