UniPROT: Uniform Prototype Selection via Partial Optimal Transport with Submodular Guarantees
About
Selecting prototypical examples from a source distribution to represent a target data distribution is a fundamental problem in machine learning. Existing subset selection methods often rely on implicit importance scores, which can be skewed towards majority classes and lead to low-quality prototypes for minority classes. We present $\methodprop$, a novel subset selection framework that minimizes the optimal transport (OT) distance between a uniformly weighted prototypical distribution and the target distribution. While intuitive, this formulation leads to a cardinality-constrained maximization of a \emph{super-additive} objective, which is generally intractable to approximate efficiently. To address this, we propose a principled reformulation of the OT marginal constraints, yielding a partial optimal transport-based submodular objective. We prove that this reformulation enables a greedy algorithm with a $(1-1/e)$ approximation guarantee relative to the original super-additive maximization problem. Empirically, we showcase that enforcing uniform prototype weights in UniPROT consistently improves minority-class representation in imbalanced classification benchmarks without compromising majority-class accuracy. In both finetuning and pretraining regimes for large language models under domain imbalance, UniPROT enforces uniform source contributions, yielding robust performance gains. Our results establish UniPROT as a scalable, theoretically grounded solution for uniform-weighted prototype selection. Our code is publicly available at GitHub\footnote{Code: https://github.com/efficiency-learning/UniPROT}
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Mathematical Reasoning | MATH | Accuracy16.8 | 882 | |
| Mathematical Reasoning | SVAMP | Accuracy86.2 | 403 | |
| Mathematical Reasoning | Mathematics | Accuracy37.2 | 46 | |
| Mathematical Reasoning | SIMULEQ | Accuracy68.28 | 30 | |
| Mathematical Reasoning | NUMGLUE | Accuracy68.8 | 30 | |
| Mathematical Reasoning | MATH | Accuracy38.4 | 26 | |
| Mathematical Reasoning | MATHINSTRUCT Evaluation Suite In-domain and Out-of-domain | GSM8K Accuracy54.89 | 11 | |
| Natural Language Understanding | SuperGLUE | Accuracy (SST2)94.65 | 6 |