AI Paper Atlas

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

Curvature-Aware Zeroth-Order Optimization for Memory-Efficient Test-Time Adaptation

Junming Zhang, Shuyu Yin, Peilin Liu, Rendong Ying, Fei Wen

CVPR 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Pseudo-label test-time adaptation · 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 curvature-aware anisotropic perturbation instead of isotropic Gaussian (ablation)ZO (k=20) · ImageNet-C sev5 (reset per corruption)average accuracy (%): 62.9 -> 69.0Improved
increase number of perturbations k from 2 to 20 (ablation)CAZO · ImageNet-C sev5 (reset per corruption)average accuracy (%): 65.2 -> 69.0Improved
increase k from 2 to 6 (ablation)CAZO · ImageNet-C Gaussian noise sev5accuracy (%): 59.3 -> 62.1Improved
increase k beyond 6 (ablation)CAZO · ImageNet-C Gaussian noise sev5accuracy and ECENo clear effect
use single-point perturbation instead of symmetric multi-point (ablation)CAZO · ImageNet-CaccuracyHurt
EMA coefficient nu 0.8 versus 1.0 (ablation)CAZO · ImageNet-C Gaussian noise sev5accuracy: 69.0Hurt
place adapter at layer 3 instead of later layers (ablation)CAZO · ImageNet-C Gaussian noise sev5accuracyImproved
use smaller adapter down-sampling ratios (96, 128) with more parameters (ablation)CAZO · ImageNet-C Gaussian noise sev5accuracyNo clear effect
swap BP-based TENT for CAZO forward-only adaptationTENT · ImageNet-C sev5memory (MB): 6,404 -> 1,695Improved
use CAZO instead of ZO under 8-bit quantizationZO · ImageNet-C sev5 (8-bit)average accuracy (%): 61.8 -> 67.8Improved
use CAZO instead of ZO under 6-bit quantizationZO · ImageNet-C sev5 (6-bit)average accuracy (%): 55.6 -> 61.2Improved
use CAZO instead of FOAFOA · ImageNet-C sev5 (reset per corruption)average accuracy (%): 65.8 -> 69.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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