We present Mode(Multi-Objective adaptive Data Efficiency), a framework that dynamically combines coreset selection strategies based on their evolving contribution to model performance. Unlike static methods, \mode adapts selection criteria to training phases: emphasizing class balance early, diversity during representation learning, and uncertainty at convergence. We show that MODE achieves (1-1/e)-approximation with O(n \log n) complexity and demonstrates competitive accuracy while providing interpretable insights into data utility evolution. Experiments show \mode reduces memory requirements
@article{arxiv.2512.21152,
title = {MODE: Multi-Objective Adaptive Coreset Selection},
author = {Tanmoy Mukherjee and Pierre Marquis and Zied Bouraoui},
journal= {arXiv preprint arXiv:2512.21152},
year = {2025}
}