Metraipolis › archive › dc_fbacaf2f8ffd

MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation — arXiv:2611.02003

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.02003
License (as archived): CC0-1.0
Archived: 2026-11-03T18:37:06.052Z
SHA-256 of the archived text: 048e77e4b2b45c5a01c1aa5628b121c39e55898ca1aa7dfa126ea3998efb6ee7
Stable URL of this page: https://metraipolis.com/documento/dc_fbacaf2f8ffd

How to cite

MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation — arXiv:2611.02003 — arXiv (en), https://arxiv.org/abs/2611.02003 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fbacaf2f8ffd (Metraipolis archive dc_fbacaf2f8ffd, sha256 048e77e4b2b45c5a01c1aa5628b121c39e55898ca1aa7dfa126ea3998efb6ee7)

BibTeX

@misc{dcfbacaf2f8ffd,
  title = {MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation — arXiv:2611.02003},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02003},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fbacaf2f8ffd; sha256 048e77e4b2b45c5a01c1aa5628b121c39e55898ca1aa7dfa126ea3998efb6ee7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02003. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Test-time adaptation (TTA) updates a deployed model using only unlabeled target data, but standard entropy-based methods often rely on deterministic confidence and can reinforce overconfident errors under distribution shift. We propose MIRA, Mutual Information guided calibration for Reliable Test-Time Adaptation, a source-free online adaptation method that uses lightweight last-layer Gaussian perturbations to probe prediction reliability. For each target batch, MIRA reuses a single feature extraction pass and samples perturbed classifiers around the source classifier, forming an efficient local predictive distribution. Rather than fitting a Bayesian posterior, MIRA treats the perturbation scale as a local probing radius and selects a batch-specific scale using mutual information (MI) and prediction agreement. The resulting perturbation-sensitivity signal identifies predictions that are confident and locally stable under classifier perturbations, while suppressing brittle predictions with high stochastic disagreement. MIRA then performs adaptation with an MI-confidence weighted entropy loss, which down-weights perturbation-sensitive or low-confidence samples before they drive entropy minimization. This weighted loss reduces the influence of brittle predictions that can otherwise cause harmful adaptation under distribution shift. The method requires no source data, retraining, model ensembles, or offline posterior fitting, and can be applied as a plug-and-play modification to entropy-minimization TTA. Experiments on distribution-shift benchmarks show improvements in accuracy, calibration, and negative log-likelihood, suggesting that mutual-information-guided perturbation probing and weighted adaptation yield more robust online TTA than deterministic confidence.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02003 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T18:12:54.534Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).