arXiv:2606. 18650v1 Announce Type: new Abstract: As Large Language Model (LLM) datasets scale to trillions of tokens, data selection has emerged as a critical frontier to filter out uninformative noise and construct adaptive learning trajectories.
The paper investigates the limits of optimal data selection for machine learning, showing that minimizing population error is impossible without a validation set much larger than the training set. It introduces a new concept called faithfulness, providing an estimable upper bound on the performance gap between training on a subset and on the full dataset. The authors propose a swap‑based local search algorithm that iteratively reduces this bound and demonstrate its effectiveness on CIFAR‑10 and CIFAR‑100 datasets.
arXiv:2209. 01754v5 Announce Type: replace-cross Abstract: The empirical risk minimization approach to data-driven decision making requires access to training data drawn under the same conditions as those that will be faced when the decision rule is deployed.
arXiv:2510. 06048v4 Announce Type: replace Abstract: Effective data selection is essential for pretraining large language models (LLMs), enhancing efficiency and improving generalization to downstream tasks.
arXiv:2602. 17894v2 Announce Type: replace-cross Abstract: Data collection is a critical component of modern statistical and machine learning pipelines, particularly when data must be gathered from multiple heterogeneous sources to study a target population of interest.
arXiv:2604. 13130v2 Announce Type: replace Abstract: We study learning to learn through the lens of hyperparameter tuning.
arXiv:2606. 01081v1 Announce Type: new Abstract: Decision-focused learning (DFL) trains predictive models by optimizing downstream decision quality rather than standalone prediction accuracy.
arXiv:2608. 11162v1 Announce Type: new Abstract: The Naive Bayes (NB) classifier remains a standard choice for categorical data, yet its widely used smoothing rules, such as Laplace, Lidstone, Krichevsky-Trofimov, and the $m$-estimate, all prescribe a fixed smoothing strength that ignores feature cardinality, sample size, and class imbalance, inducing a non-vanishing bias on modern high-cardinality tabular data.
arXiv:2602. 02877v2 Announce Type: replace Abstract: This paper studies optimization for a family of problems termed $\textbf{compositional entropic risk minimization}$, in which each data's loss is formulated as a Log-Expectation-Exponential (Log-E-Exp) function.
FedFIbOS introduces a Fisher‑importance based criterion for selecting submodel parameters in heterogeneous federated learning, addressing the lack of theoretical justification in prior heuristic methods. By deriving a Fisher‑weighted quadratic masking surrogate and showing that the raw Fisher top‑k rule satisfies this surrogate under a Fisher‑dominant ranking condition, the method preserves convergence guarantees while efficiently estimating Fisher scores from squared gradients. Experiments on CIFAR‑10, CIFAR‑100, and AGNews demonstrate that FedFIbOS outperforms state‑of‑the‑art approaches by roughly 10% in accuracy, especially under strong non‑IID heterogeneity.
arXiv:2606. 14971v1 Announce Type: cross Abstract: While large and diverse datasets have driven recent advances in large models, identifying the optimal data mixture for pre-training and post-training remains a significant open problem.
arXiv:2506. 11336v2 Announce Type: replace Abstract: We study the sample complexity of stochastic convex optimization when problem parameters such as the distance to optimality and the Lipschitz constant are unknown.