AI Paper Atlas

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

TTL: Test-time Textual Learning for OOD Detection with Pretrained Vision-Language Models

Jinlun Ye, Jiang Liao, Runhe Lai, Xinhua Lu, Jiaxin Zhuang, Zhiyong Gan, Ruixuan Wang

CVPR 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 L_OMB pseudo-label prompt learning (ablation)MCM base detector · ImageNet-1kFPR95: 42.77 -> 30.56Improved
add L_OMB pseudo-label prompt learning (ablation)MCM base detector · CIFAR-100FPR95: 73.09 -> 14.40Improved
add OOD knowledge purification L_OKP (ablation)TTL (L_OMB + OKB) · ImageNet-1kFPR95: 18.40 -> 12.46Improved
add OOD knowledge purification L_OKP (ablation)TTL (L_OMB + OKB) · CIFAR-100FPR95: 5.23 -> 2.36Improved
add OOD Textual Knowledge Bank (ablation)TTL (L_OMB + L_OKP) · ImageNet-1kFPR95: 24.59 -> 12.46Improved
add OOD Textual Knowledge Bank (ablation)TTL (L_OMB + L_OKP) · CIFAR-100FPR95: 4.14 -> 2.36Improved
use proposed lowest-similarity OKB update instead of FIFO (ablation)TTL · ImageNet-1kFPR95: 14.69 -> 12.46Improved
use proposed OKB update instead of random replacement (ablation)TTL · CIFAR-100FPR95: 29.06 -> 2.36Improved
initialize OOD prompt prefix and classname manually instead of randomly (ablation)TTL · CIFAR-100FPR95: 45.23 -> 2.36Improved
initialize prefix with 'a photo of a' while keeping random classname (ablation)TTL · CIFAR-100FPR95: 45.23 -> 45.29No clear effect
add a visual knowledge bank (TTL-V)TTL · ImageNet-1kFPR95: 12.46 -> 9.19Improved
stop OOD prompt updates after 1,280 test samples (ablation)TTL · ImageNet-1kFPR95: 12.46 -> 14.21Hurt

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