# Not Every Correction Helps: Gain-Guided Continual Test-Time Adaptation

arXiv: https://arxiv.org/abs/2609.36655 · HTML: https://mlsanity.com/paper/2609.36655/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 44 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 30 of 36)

- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Learning Robust Global Representations by Penalizing Local Predictive Power](https://mlsanity.com/paper/1905.13549/) (2019)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [Robust Test-Time Adaptation in Dynamic Scenarios](https://mlsanity.com/paper/2303.13899/) (2023)
- [A General Framework for Updating Belief Distributions](https://mlsanity.com/paper/1306.6430/) (2013)
- [EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization](https://mlsanity.com/paper/2303.01904/) (2023)
- [3D Common Corruptions and Data Augmentation](https://mlsanity.com/paper/2203.01441/) (2022)
- [Entropy is not Enough for Test-Time Adaptation: From the Perspective of Disentangled Factors](https://mlsanity.com/paper/2403.07366/) (2024)
- [Test-Time Model Adaptation with Only Forward Passes](https://mlsanity.com/paper/2404.01650/) (2024)
- [Universal Test-time Adaptation through Weight Ensembling, Diversity Weighting, and Prior Correction](https://mlsanity.com/paper/2306.00650/) (2023)
- [SoTTA: Robust Test-Time Adaptation on Noisy Data Streams](https://mlsanity.com/paper/2310.10074/) (2023)
- [ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation](https://mlsanity.com/paper/2306.04344/) (2023)
- [A Probabilistic Framework for Lifelong Test-Time Adaptation](https://mlsanity.com/paper/2212.09713/) (2022)
- [Continual Test-time Domain Adaptation via Dynamic Sample Selection](https://mlsanity.com/paper/2310.03335/) (2023)
- [DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation](https://mlsanity.com/paper/2406.10737/) (2024)
- [Continual-MAE: Adaptive Distribution Masked Autoencoders for Continual Test-Time Adaptation](https://mlsanity.com/paper/2312.12480/) (2023)
- [BECoTTA: Input-dependent Online Blending of Experts for Continual Test-time Adaptation](https://mlsanity.com/paper/2402.08712/) (2024)
- [DOTA: Distributional Test-Time Adaptation of Vision-Language Models](https://mlsanity.com/paper/2409.19375/) (2024)
- [Ranked Entropy Minimization for Continual Test-Time Adaptation](https://mlsanity.com/paper/2505.16441/) (2025)
- [Realistic Test-Time Adaptation of Vision-Language Models](https://mlsanity.com/paper/2501.03729/) (2025)
- [Beyond Entropy: Region Confidence Proxy for Wild Test-Time Adaptation](https://mlsanity.com/paper/2505.20704/) (2025)
- [NEO: No-Optimization Test-Time Adaptation through Latent Re-Centering](https://mlsanity.com/paper/2510.05635/) (2025)
- [BayesTTA: Continual-Temporal Test-Time Adaptation for Vision-Language Models via Gaussian Discriminant Analysis](https://mlsanity.com/paper/2507.08607/) (2025)
- [Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment](https://mlsanity.com/paper/2508.15568/) (2025)
- [C-PMI: Conditional Pointwise Mutual Information for Turn-level Dialogue Evaluation](https://mlsanity.com/paper/2306.15245/) (2023)
- [Analytic Continual Test-Time Adaptation for Multi-Modality Corruption](https://mlsanity.com/paper/2410.22373/) (2024)

## Abstract (arXiv)

Continual test-time adaptation (CTTA) adapts a source model to an unlabeled test stream whose distribution may change over time. Existing TTA methods often assess prediction reliability using confidence or entropy, which primarily reflect the model's self-certainty for the current sample. In CTTA, accumulated target observations can provide complementary evidence for correcting the source prediction, but this history may become misaligned as the target distribution changes. The key question is therefore not how much the correction differs from the source prediction, but whether and how strongly it should be applied. This paper proposes Gain-Aware INtervention (GAIN), a backpropagation-free CTTA framework guided by a simple principle: history proposes, gain decides. GAIN maintains compact target statistics to form a correction proposal and a posterior-predictive evaluator that accounts for estimation uncertainty. The resulting source-relative gain estimates the proposal's benefit and determines a sample-specific intervention strength along a continuous path through efficient one-dimensional optimization. Gain-controlled predictions then update the target statistics online, limiting the propagation of unreliable corrections, all without backpropagation, sample storage, or replay. Across five benchmarks, our method achieves strong predictive performance, with favorable accuracy--calibration--efficiency trade-offs in continual adaptation. On ImageNet-C, for example, GAIN achieves 61.9% accuracy with near-source calibration. It remains stable under diverse and challenging continual shifts while running 15.9x faster than a representative optimization-based CTTA baseline.