Papers › Annotation-Efficient Active Test-Time Adaptation with Conformal Prediction
Annotation-Efficient Active Test-Time Adaptation with Conformal Prediction
Tingyu Shi, Fan Lyu, Shaoliang Peng
Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).
Active Test-Time Adaptation (ATTA) improves model robustness under domain shift by selectively querying human annotations at deployment, but existing methods use heuristic uncertainty measures and suffer from low data selection efficiency, wasting human annotation budget. We propose Conformal Prediction Active TTA (CPATTA), which first brings principled, coverage-guaranteed uncertainty into ATTA. CPATTA employs smoothed conformal scores with a top-K certainty measure, an online weight-update algorithm driven by pseudo coverage, a domain-shift detector that adapts human supervision, and a staged update scheme balances human-labeled and model-labeled data. Extensive experiments demonstrate that CPATTA consistently outperforms the state-of-the-art ATTA methods by around 5% in accuracy. Our code and datasets are available at https://github.com/tingyushi/CPATTA.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Syntology holds the repository link but has not harvested or run code from it.
Results from the paper
The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2509.25692, was issued in September 2025, after that date, so the archive has no leaderboard rows for it.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections