# Improving Model Classification by Optimizing the Training Dataset

arXiv: https://arxiv.org/abs/2507.16729 · HTML: https://mlsanity.com/paper/2507.16729/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 21 references that are in the map.
- First posted on arXiv 22 Jul 2025.

## Builds on (most cited first, 11 of 11)

- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [On Coresets for Logistic Regression](https://mlsanity.com/paper/1805.08571/) (2018)
- [Coresets for Near-Convex Functions](https://mlsanity.com/paper/2006.05482/) (2020)
- [Coreset-Based Neural Network Compression](https://mlsanity.com/paper/1807.09810/) (2018)
- [Pruning Neural Networks via Coresets and Convex Geometry: Towards No Assumptions](https://mlsanity.com/paper/2209.08554/) (2022)
- [On Coresets for Support Vector Machines](https://mlsanity.com/paper/2002.06469/) (2020)
- [Sets Clustering](https://mlsanity.com/paper/2003.04135/) (2020)
- [The Power of Uniform Sampling for Coresets](https://mlsanity.com/paper/2209.01901/) (2022)
- [New Coresets for Projective Clustering and Applications](https://mlsanity.com/paper/2203.04370/) (2022)
- [Coresets for Decision Trees of Signals](https://mlsanity.com/paper/2110.03195/) (2021)
- [An Efficient Drifters Deployment Strategy to Evaluate Water Current Velocity Fields](https://mlsanity.com/paper/2301.04216/) (2023)

## Built on by (most cited first, 1 of 1)

- [ARIA: Adaptive Region-Based Importance Allocation for Conditional Diffusion Distillation](https://mlsanity.com/paper/2606.23898/) (2026)

## Abstract (arXiv)

In the era of data-centric AI, the ability to curate high-quality training data is as crucial as model design. Coresets offer a principled approach to data reduction, enabling efficient learning on large datasets through importance sampling. However, conventional sensitivity-based coreset construction often falls short in optimizing for classification performance metrics, e.g., F1 score, focusing instead on loss approximation. In this work, we present a systematic framework for tuning the coreset generation process to enhance downstream classification quality. Our method introduces new tunable parameters--including deterministic sampling, class-wise allocation, and refinement via active sampling, beyond traditional sensitivity scores. Through extensive experiments on diverse datasets and classifiers, we demonstrate that tuned coresets can significantly outperform both vanilla coresets and full dataset training on key classification metrics, offering an effective path towards better and more efficient model training.