Haokun Lin
3
Papers
473
Total Citations
Papers (3)
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
ECCV 2024arXiv
473
citations
DOGR: Towards Versatile Visual Document Grounding and Referring
ICCV 2025
0
citations
MoPE-CLIP: Structured Pruning for Efficient Vision-Language Models with Module-wise Pruning Error Metric
CVPR 2024
0
citations