4 papers
Respect Your Zero-Shot Uncertainty: Conservative Calibration for Test-Time-Adapted Vision-Language Models
Jingyan Jiang, Yaru Sun, Xiao Chen +5
Test-time adaptation (TTA) can improve the recognition accuracy of vision-language models under distribution shift, but often degrades calibration, making predictive confidence unr…
Von Mises-Fisher Mixture Model with Dynamic Shrinkage for Realistic Test-Time Transduction
Jiazhen Huang, Zhiming Liu, Changhu Wang +3
A range of methods aim to enhance the performance of vision-language models (VLMs) at test time. Among them, transduction has emerged as a promising paradigm due to its strong comp…
What Drives Test-Time Adaptation for CLIP? A Controlled Empirical Study from an Update Perspective
Jiazhen Huang, Xiao Chen, Zhiming Liu +3
Vision-Language Models (VLMs) such as CLIP have become a standard backbone for open-vocabulary recognition, yet their zero-shot predictions remain vulnerable to distribution shifts…
Skill-Conditioned Gated Self-Distillation for LLM Reasoning
Jiazhen Huang, Xiao Chen, Xiao Luo +3
On-policy self-distillation (SD) improves LLM reasoning by using teacher-side privileged information (PI) to turn sparse verifier outcomes into dense token-level supervision. Exist…