Xiangyu Wu

Regression-based Test-Time Adaptation of Vision-Language Models

Multimodal Classification via Total Correlation Maximization

Adaptive Debiasing Tsallis Entropy for Test-Time Adaptation

LPO: Towards Accurate GUI Agent Interaction via Location Preference Optimization

MirrorCAPTCHA: Wild CAPTCHA, Wild Distribution, Wild Web-based Platform Meet Multimodal LLM Agents

Multi-Label Test-Time Adaptation with Bound Entropy Minimization

Text as Any-Modality for Zero-shot Classification by Consistent Prompt Tuning