7 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…
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…
Test-Time Distillation for Continual Model Adaptation
Xiao Chen, Jiazhen Huang, Zhiming Liu +4
Deep neural networks often suffer performance degradation upon deployment due to distribution shifts. Continual Test-Time Adaptation (CTTA) aims to address this issue in an unsuper…
Semantic-Space Exploration and Exploitation in RLVR for LLM Reasoning
Fanding Huang, Guanbo Huang, Xiao Fan +7
Reinforcement Learning with Verifiable Rewards (RLVR) for LLM reasoning is often framed as balancing exploration and exploitation in action space, typically operationalized with to…
Neural Collapse in Test-Time Adaptation
Xiao Chen, Zhongjing Du, Jiazhen Huang +4
Test-Time Adaptation (TTA) enhances model robustness to out-of-distribution (OOD) data by updating the model online during inference, yet existing methods lack theoretical insights…