AI RESEARCH
ProtoDCS: Towards Robust and Efficient Open-Set Test-Time Adaptation for Vision-Language Models
arXiv CS.AI
•
ArXi:2602.23653v2 Announce Type: replace-cross Large-scale Vision-Language Models (VLMs) exhibit strong zero-shot recognition, yet their real-world deployment is challenged by distribution shifts. While Test-Time Adaptation (TTA) can mitigate this, existing VLM-based TTA methods operate under a closed-set assumption, failing in open-set scenarios where test streams contain both covariate-shifted in-distribution (csID) and out-of-distribution (csOOD) data.