Author

Guangyi Yang

h-index 14 View on Semantic Scholar

We have 1 of 47 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

DIVE: Dynamic Iterative Visual Evidence Construction for Efficient Vision-Language Models

DIVE (Dynamic Iterative Visual Evidence Construction), a training-free framework that recasts visual-token pruning as dynamic evidence construction and builds a retained set of complementary, prompt-relevant evidence.

Cheng Zhong, Xiao An, Zijie Wang et al. · 0 citations