REVIEW 14 cited by
Beyond Model Adaptation at Test Time: A Survey
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Machine learning algorithms have achieved remarkable success across various disciplines, use cases and applications, under the prevailing assumption that training and test samples are drawn from the same distribution. Consequently, these algorithms struggle and become brittle even when samples in the test distribution start to deviate from the ones observed during training. Domain adaptation and domain generalization have been studied extensively as approaches to address distribution shifts across test and train domains, but each has its limitations. Test-time adaptation, a recently emerging learning paradigm, combines the benefits of domain adaptation and domain generalization by training models only on source data and adapting them to target data during test-time inference. In this survey, we provide a comprehensive and systematic review on test-time adaptation, covering more than 400 recent papers. We structure our review by categorizing existing methods into five distinct categories based on what component of the method is adjusted for test-time adaptation: the model, the inference, the normalization, the sample, or the prompt, providing detailed analysis of each. We further discuss the various preparation and adaptation settings for methods within these categories, offering deeper insights into the effective deployment for the evaluation of distribution shifts and their real-world application in understanding images, video and 3D, as well as modalities beyond vision. We close the survey with an outlook on emerging research opportunities for test-time adaptation.
Forward citations
Cited by 14 Pith papers
-
Respect Your Zero-Shot Uncertainty: Conservative Calibration for Test-Time-Adapted Vision-Language Models
ZAEC anchors calibration to each sample's zero-shot entropy and selectively softens over-sharpened TTA predictions, reaching the lowest macro-average calibration error among evaluated post-hoc methods on ViT-B/16.
-
Family Matters: A Systematic Study of Spatial vs. Frequency Masking for Continual Test-Time Adaptation
Random spatial patch masking keeps continual test-time adaptation stable on long corrupted streams with ViTs, whereas random frequency-band masking collapses; the gap shrinks on CNNs and on global-cue tasks with large ViTs.
-
Active Test-time Vision-Language Navigation
ATENA uses episodic success/failure labels and a mixture entropy objective to adapt vision-language navigation policies at test time, improving REVERIE, R2R, and R2R-CE benchmarks.
-
Noise is an Efficient Learner for Zero-Shot Vision-Language Models
Optimizing a clamped Gaussian noise map on the input image at test time with entropy and inter-view consistency losses improves zero-shot CLIP accuracy on natural distribution shifts.
-
Chaos Is a LADDER: Domain Generalization Beyond Invariance via Reweighting
Reweighting frozen source classifiers by target style-distribution distance (Sinkhorn-KNN) beats pooled and invariant predictors on some rule-varying DG benchmarks.
-
Test-Time Adaptation via Dual Distillation for Videos Under Severe Distribution Shifts
Online dual distillation of a lightweight CLIP adapter (zero-shot prior plus frozen source adapter) yields state-of-the-art video test-time adaptation under severe natural domain shifts.
-
Continual Test-Time Adaptation in Computer Vision: Methods, Benchmarks, and Future Directions
CTTA methods fall into optimization-based, parameter-efficient, and architecture-based families that adapt pretrained vision models online under continual unlabeled shifts while fighting forgetting and error accumulation.
-
Backpropagation-Free Test-Time Adaptation for Lightweight EEG-Based Brain-Computer Interfaces
A backpropagation-free test-time adaptation method, BFT, weights predictions from multiple transformed copies of each EEG test sample using a learning-to-rank module, improving cross-subject decoding without model updates.
-
Adapting Vision-Language Models Without Labels: A Comprehensive Survey
A survey that organizes unsupervised vision-language model adaptation by unlabeled-data availability into four paradigms: data-free transfer, domain transfer, episodic test-time, and online test-time adaptation.
-
Latte: Collaborative Test-Time Adaptation of Vision-Language Models in Federated Learning
Latte improves federated test-time adaptation of CLIP by merging each client's local memory with prototypes retrieved from similar clients.
-
Visual Instance-aware Prompt Tuning
ViaPT generates instance-aware prompts per image, fuses them with dataset-level prompts, and applies PCA compression to outperform VPT-Deep and other PEFT baselines on FGVC, HTA, and VTAB-1k.
-
PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation
PAID proposes Householder-based orthogonal weight updates for continual test-time adaptation, claiming that preserving pairwise angular structure of pretrained weights is a useful prior, but the math and validation fo...
-
No Training, Better Flights: Test-Time Scaled VLMs for UAV Navigation
Test-time scaling—parallel candidate generation, iterative self-correction, and multi-criteria selection—improves a frozen UAV navigation VLM's success rate by about 2 percentage points on the TravelUAV benchmark.
-
Continuous Self-Improvement of Large Language Models by Test-time Training with Verifier-Driven Sample Selection
A test-time training method that fine-tunes LoRA adapters on verifier-selected high-confidence pseudo-labels, reporting large gains on math benchmarks, but evaluated on the same queries it adapts on.
Discussion (0). Continue with ORCID to comment.