AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Probe, Anchor, and Amend: Active Test-Time Adaptation of Vision-Language Models

Jiaqi Lin, Chaoqi Chen, Xiasi Wang, Mingfu Yan, Jiancheng Huang, Jianzhuang Liu, Wenming Yang, Qingmin Liao

ECCV 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add alignment loss L_align (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.80 -> 68.19Improved
add intra-class compactness loss L_intra (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.92 -> 68.19Improved
add inter-class separation loss L_inter (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 68.03 -> 68.19Improved
add memory-driven amend loss L_amend (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.98 -> 68.19Improved
train visual LayerNorm parameters instead of freezing (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.28 -> 68.19Improved
train textual residuals instead of freezing (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.38 -> 68.19Improved
use textual residuals instead of text-side LayerNorm (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 67.61 -> 68.19Improved
increase prototype cache size from 1 to 3 (ablation)PAA · ImageNet-A, ViT-B/16top-1 accuracy (%): 65.22 -> 66.59Improved
increase prototype cache size from 3 to 7 (ablation)PAA · ImageNet-A, ViT-B/16top-1 accuracy (%): 66.59 -> 65.52Hurt
increase memory lifetime from 1 to 3 (ablation)PAA · ImageNet-A, ViT-B/16top-1 accuracy (%): 66.32 -> 66.59Improved
increase memory lifetime from 3 to 5 (ablation)PAA · ImageNet-A, ViT-B/16top-1 accuracy (%): 66.59 -> 66.63No clear effect
use full-class labels instead of binary verification (ablation)PAA · Natural Distribution Shifts avg, ViT-B/16top-1 accuracy (%): 68.19 -> 68.84Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy