Author

Wei-Zhen Zhou

h-index 0 View on Semantic Scholar

We have 1 of 1 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 Sep 2026

Incentive Noise and Structural Prior Infusion for Multi-modal Object Re-Identification

This work proposes a novel framework centered on Positive-Incentive Noise and structured prompt modulation that harnesses task-aware {\pi}-noise, sampled from a distribution conditioned on both visual and text inputs, to perturb global tokens and enable semantics-guided cross-modal compensation.

Wei-Xiang Zhou, Yu-Hao Wang, Xing-Guo Xu et al. · 0 citations