AI Paper Atlas

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

Stabilizing Real-World Visual Active Tracking with Action-Smooth Test-Time Adaptation

Haowei Sun, Shiteng Zhang, Jinwu Hu, Kaining Chen, Mingkui Tan

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

Read the paper

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 Critic Value Maximization (ablation)VATA (EVT base) · UnrealCV averageSR: 0.80 -> 0.85Improved
add Action Smoothness Regularization (ablation)VATA (EVT base) · UnrealCV averageSR: 0.82 -> 0.85Improved
apply VATA test-time adaptation to base trackerEVT · UnrealCV averageSR: 0.79 -> 0.85Improved
use VATA instead of entropy-based TARLVATA vs TARL · UnrealCV averageSR: 0.82 -> 0.85Improved
apply VATA (critic value maximization only, discrete actions)GC-VAT · DAT averageTSR: 0.62 -> 0.67Improved
use critic value maximization instead of entropy minimization (TARL)VATA vs TARL · DAT averageTSR: 0.64 -> 0.67Improved
apply VATA to base trackerEVT · EVT-Bench averageSR: 15.1% -> 22.4%Improved
apply VATA versus runner-up TTA on distractor taskVATA vs TARL · EVT-Bench DTSR: 6.8% -> 11.2%Improved
add action smoothness regularization to imitation-learning trackerOpen-TrackVLA · EVT-Bench DTSR: 11.6% -> 16.9%Improved
deploy VATA on a real droneEVT · Real drone (10 episodes)TSR: 0.2 -> 0.8Improved
vary critic threshold Q0 in {5,10,20} and jump threshold delta in {0.1,0.2,0.3} (ablation)VATA · UnrealCV averageAR: AR variation less than 3.3%No clear effect
filter samples by critic value rather than output entropy (ablation)VATA · DAT (GC-VAT)success vs loss gap: critic 9.12 -> 3.10 on target loss; entropy 0.580 vs 0.560Improved

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