PTTA: Purifying Malicious Samples for Test-Time Model Adaptation

Jing Ma, Hanlin Li, Xiang Xiang
2/14/2026

No official code found. Best implementation: HAIV-Lab/ptta (8 stars, last commit 2025-10-27).

Semantic Scholar

Code Implementations(1)

HAIV-Lab/pttaCommunity71%

[ICML 2025] PTTA: Purifying Malicious Samples for Test-Time Model Adaptation

80Shell, PythonMay 28, 202511 months ago

Cite this paper

@article{ma2026ptta,
  title  = {PTTA: Purifying Malicious Samples for Test-Time Model Adaptation},
  author = {Jing Ma and Hanlin Li and Xiang Xiang},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:81271dfaa67a77b3a98f32dbc583bad5bc138959},
  journal = {ICML 2025 2025}
}

Discussion