arXiv:2606. 15569v1 Announce Type: new Abstract: Test-time training (TTT) adapts a pretrained model to each prompt via parameter updates, improving accuracy under pretraining-to-test distribution shifts.
arXiv:2609.36655v1 Announce Type: new Abstract: Continual test-time adaptation (CTTA) adapts a source model to an unlabeled test stream whose distribution may change over time. Existing TTA methods o...
arXiv:2601. 07094v2 Announce Type: replace-cross Abstract: Bayesian optimization (BO) iteratively fits a Gaussian process (GP) surrogate to accumulated evaluations and selects new queries via an acquisition function.
arXiv:2609.37687v1 Announce Type: new Abstract: Active test-time adaptation (ATTA) improves robustness under distribution shift by updating a deployed model during inference while selectively queryin...
arXiv:2607. 00259v1 Announce Type: cross Abstract: Test-Time Adaptation (TTA) seeks to improve model robustness under distribution shifts by adapting parameters using unlabeled target data.
The paper studies how test‑time training (TTT) improves in‑context learning (ICL) for nonlinear models, focusing on single‑index models where features lie in a hidden low‑dimensional subspace. By applying TTT to single‑layer transformers trained with gradient‑based methods, the authors derive an upper bound on prediction risk and show that TTT allows the model to adapt to both feature vectors and link functions that vary across tasks—something ICL alone struggles to achieve. They also provide a convergence rate indicating that predictive error can approach the noise level as context size and network width increase.
The paper introduces a statistical framework for post‑training hyperparameter selection, emphasizing the learn‑then‑test (LTT) paradigm. It treats hyperparameter tuning as a multiple hypothesis testing problem over a candidate set, enabling the selection of hyperparameters that meet specified reliability constraints such as risk bounds or information‑theoretic limits. The framework provides finite‑sample control of error probabilities using p‑values, e‑values, and concentration inequalities derived from first principles.
arXiv:2609. 11235v1 Announce Type: new Abstract: Test-time adaptation (TTA) offers many ways to update a deployed model without labels, but choosing the wrong update can make a strong source model worse.
The paper investigates the teacher‑student framework used in Test‑Time Adaptation (TTA) and questions the common practice of updating the teacher via an exponential moving average of the student. The authors demonstrate that error accumulation still occurs, especially over longer sequences, and propose an intransigent teacher that remains fixed. This modification yields significant performance gains across multiple datasets, longer scenarios, and various architectures, including semantic segmentation, while also improving robustness to hyperparameter changes.
arXiv:2608.20998v1 Announce Type: new Abstract: Reservoir computing (RC) couples a fixed recurrent dynamical system with a trained lightweight readout, but this efficiency is partly lost during hyper...
The paper introduces KENDO, a unified framework that combines Ensemble Gaussian Processes with disagreement‑aware acquisition strategies to address hyperparameter selection in Bayesian optimization and active learning. By replacing costly hyperparameter sampling with a kernel ensemble and adaptive Bayesian weighting, KENDO‑BO and KENDO‑AL provide self‑correcting mechanisms tailored to their respective tasks. Experiments on synthetic and real‑world benchmarks show that KENDO‑BO matches or outperforms state‑of‑the‑art methods while cutting computational cost up to fivefold, and KENDO‑AL delivers better predictive calibration with up to 27‑times speedup compared to MCMC‑based baselines.
arXiv:2607. 09415v1 Announce Type: cross Abstract: Long-context processing has become increasingly important for large language models (LLMs), but simply extending the context window does not guarantee effective utilization of long inputs.
arXiv:2605. 28057v2 Announce Type: replace-cross Abstract: Test-time adaptation (TTA) aims to adapt models to maintain reliable performance on non-stationary test streams without requiring labeled data.