Pith. sign in

REVIEW 24 cited by

Towards Stable Test-Time Adaptation in Dynamic Wild World

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

arxiv 2302.12400 v1 pith:NOBOVIT5 submitted 2023-02-24 cs.LG cs.CV

Shuaicheng Niu , Jiaxiang Wu , Yifan Zhang , Zhiquan Wen , Yaofo Chen , Peilin Zhao , Mingkui Tan This is my paper · ORCID
classification cs.LGcs.CV
keywords modelsamplestestdistributionlayernoisynormshifts
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Test-time adaptation (TTA) has shown to be effective at tackling distribution shifts between training and testing data by adapting a given model on test samples. However, the online model updating of TTA may be unstable and this is often a key obstacle preventing existing TTA methods from being deployed in the real world. Specifically, TTA may fail to improve or even harm the model performance when test data have: 1) mixed distribution shifts, 2) small batch sizes, and 3) online imbalanced label distribution shifts, which are quite common in practice. In this paper, we investigate the unstable reasons and find that the batch norm layer is a crucial factor hindering TTA stability. Conversely, TTA can perform more stably with batch-agnostic norm layers, \ie, group or layer norm. However, we observe that TTA with group and layer norms does not always succeed and still suffers many failure cases. By digging into the failure cases, we find that certain noisy test samples with large gradients may disturb the model adaption and result in collapsed trivial solutions, \ie, assigning the same class label for all samples. To address the above collapse issue, we propose a sharpness-aware and reliable entropy minimization method, called SAR, for further stabilizing TTA from two aspects: 1) remove partial noisy samples with large gradients, 2) encourage model weights to go to a flat minimum so that the model is robust to the remaining noisy samples. Promising results demonstrate that SAR performs more stably over prior methods and is computationally efficient under the above wild test scenarios.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 24 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 63 citations worldwide. Full citation record

  1. eSkinHealth: A Multimodal Dataset for Neglected Tropical Skin Diseases

    cs.AI 2025-08 conditional novelty 7.0 of 10

    eSkinHealth is a new West African skin-disease dataset with 5,623 images, 47 conditions, and multimodal annotations including masks, captions, and clinical concepts for AI dermatology research.

  2. EmbodiTTA: Resource-Efficient Test-Time Adaptation for Embodied Visual Systems

    cs.LG 2025-05 unverdicted novelty 7.0 of 10

    OD-TTA enables resource-efficient test-time adaptation on edge devices by triggering updates only on detected domain shifts, achieving comparable accuracy with lower energy and computation costs for embodied visual systems.

  3. Suppress and Diversify: Refining Robust Pathways for Corruption Robustness

    cs.CV 2026-08 conditional novelty 6.0 of 10

    S&D improves corruption robustness by selecting the most stable internal pathways under a synthetic corruption and diversifying them through symmetric weight tweaks, with no test-time overhead.

  4. Respect Your Zero-Shot Uncertainty: Conservative Calibration for Test-Time-Adapted Vision-Language Models

    cs.CV 2026-08 conditional novelty 6.0 of 10

    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.

  5. Test-Time Curriculum for Open-Set AIGC Detection

    cs.CV 2026-08 conditional novelty 6.0 of 10

    A curriculum-based test-time adaptation method, using balanced confident pseudo-labels and multi-scale refinement, improves AIGC detector accuracy on unseen generators by 11 to 29 points over its starting detector.

  6. When Can Test-Time Adaptation Help Zero-Shot CT Vision-Language Models?

    cs.CV 2026-07 conditional novelty 6.0 of 10

    CARVE is a label-free, cardinality-aware test-time adaptation method that consistently improves multi-label CT diagnosis when the base model is already discriminative and input depth matches pretraining.

  7. No Time Like the Present: Agentic Test-Time Training for LLM Agents

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Continuous test-time training that downweights tokens in repeated n-grams improves multi-turn LLM agent success by up to about 5 points by limiting self-reinforcing drift.

  8. When Small Guides Large: Cross-Model Co-Learning for Test-Time Adaptation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    COCA shows that bidirectional cross-model co-learning during test-time adaptation lets even small models significantly improve large models' robustness to distribution shift.

  9. Align the GAP: Prior-based Unified Multi-Task Remote Physiological Measurement Framework For Domain Generalization and Personalization

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A prior-based unified framework, GAP, jointly handles multi-source domain generalization and per-user test-time adaptation for multi-task rPPG, beating prior domain generalization and test-time adaptation methods on s...

  10. Frequency Composition for Compressed and Domain-Adaptive Neural Networks

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Training quantized models on low-frequency images plus frequency-aware batch normalization at test time improves both compression and domain-shift robustness.

  11. Learning Compositional Transferability of Time Series for Source-Free Domain Adaptation

    cs.LG 2025-04 conditional novelty 6.0 of 10

    A compositional U-net-plus-autoencoder reconstruction scheme with test-time stability rescaling reports state-of-the-art MF1 scores on three time series source-free domain adaptation benchmarks.

  12. Improving Batch Normalization with TTA for Robust Object Detection in Self-Driving

    cs.CV 2024-11 conditional novelty 6.0 of 10

    A per-layer learnable BN mixing coefficient, optimized at test time with an entropy plus max-min loss, improves BEV-based 3D detection on some NuScenes-C corruptions by up to ~8% NDS.

  13. Robustness Emerges Early in Training Dynamics, but Is Not Preserved

    cs.LG 2026-08 conditional novelty 5.0 of 10

    Shallow layers are most robust to corruptions early in training; freezing them (EPS) or rewinding them (AWR) reduces corruption error on several benchmarks, with caveats.

  14. SloMo-Fast: Slow-Momentum and Fast-Adaptive Teachers for Source-Free Continual Test-Time Adaptation

    cs.LG 2025-11 reject novelty 5.0 of 10

    SloMo-Fast, a source-free dual-teacher CTTA method with test-time prototypes, reports SOTA error rates across 11 settings and a new cyclic domain-repetition benchmark, but the appendix data contain serious inconsistencies.

  15. DCFS: Continual Test-Time Adaptation via Dual Consistency of Feature and Sample

    cs.CV 2025-08 conditional novelty 5.0 of 10

    A new continual test-time adaptation framework that disentangles target features into semantic and domain parts, enforces dual consistency, and weights self-supervised losses by confidence, achieving small average err...

  16. Enhancing Novel View Synthesis from extremely sparse views with SfM-free 3D Gaussian Splatting Framework

    cs.CV 2025-08 unverdicted novelty 5.0 of 10

    SfM-free 3D Gaussian Splatting jointly estimates camera poses and geometry from two views and reports a 2.75 dB PSNR improvement over existing 3DGS methods on extremely sparse inputs.

  17. Continual Adaptation: Environment-Conditional Parameter Generation for Object Detection in Dynamic Scenarios

    cs.CV 2025-06 conditional novelty 5.0 of 10

    A conditional diffusion model generates LoRA adapter parameters for object detectors at test time, improving continual domain adaptation accuracy by small margins over prior methods.

  18. Single Image Test-Time Adaptation via Multi-View Co-Training

    cs.CV 2025-06 conditional novelty 5.0 of 10

    MuVi achieves near-supervised breast MRI tumor segmentation by adapting a pretrained 3D model on a single test image via multi-view co-training with entropy-thresholded pseudolabels.

  19. PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation

    cs.CV 2025-06 reject novelty 5.0 of 10

    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...

  20. TTAQ: Towards Stable Post-training Quantization in Continuous Domain Adaptation

    cs.LG 2024-12 conditional novelty 5.0 of 10

    TTAQ combines weight standardization, perturbation consistency reconstruction, and adaptive balanced loss to improve post-training quantized models under continual test-time domain shift.

  21. Scalable Out-of-distribution Robustness in the Presence of Unobserved Confounders

    cs.LG 2024-11 conditional novelty 5.0 of 10

    Under weak overlap, a latent confounder can be approximately identified from a single proxy or multiple unlabeled sources, and a reweighted mixture-of-experts model adapts to confounder shift.

  22. Billet Number Recognition Based on Test-Time Adaptation

    cs.CV 2025-02 conditional novelty 4.0 of 10

    A test-time adaptation plus prior-knowledge correction method raises billet number recognition accuracy from 58% to 80% on machine-printed and from 24% to 70% on handwritten samples in the authors' tests.

  23. SPARNet: Continual Test-Time Adaptation via Sample Partitioning Strategy and Anti-Forgetting Regularization

    cs.CV 2025-01 reject novelty 4.0 of 10

    SPARNet combines sample partitioning, generalized entropy minimization, mean-teacher consistency, and an anti-forgetting regularizer to improve continual test-time adaptation on CIFAR10-C, CIFAR100-C, and ImageNet-C.

  24. Generalizing vision-language models to novel domains: A comprehensive survey

    cs.CV 2025-06 conditional novelty 3.0 of 10

    A survey of VLM generalization literature organized by transferred module, with benchmark tables and a review of multimodal LLMs.

Pith tools