Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

HARP: Efficient Data Selection for Finetuning Large Language Models

About

Finetuning data selection requires balancing two competing goals: selecting examples that improve the downstream objective, and doing so without repeatedly finetuning models. Train-free selectors are scalable but rely on proxies such as embedding similarity or clustering, which may not match the target objective. Train-based selectors better reflect downstream utility through gradient signals, subset evaluation, or Shapley attribution, but require many costly train--evaluate iterations. We propose Hierarchical Active Region Pruning (HARP), an efficient train-based selector that preserves downstream alignment while reducing selection cost. HARP organizes the training pool into a node--leaf hierarchy, evaluates only representative leaves, and infers unmeasured utilities with empirical Bayes posteriors. It then selects data using two complementary envelopes: HARP-C, which conservatively controls redundancy, and HARP-E, which additively rewards complementary regions. We theoretically show that, under local smoothness and bounded estimation error, HARP controls selection error while reducing train--evaluate cost. We further validate that HARP variants achieve the best result and outperform the strongest baseline by up to $+8.9$ points, while using roughly $7\times$ fewer training examples.

Ning Wang, Zhengxin Zhang, Maosen Tang, Yitang Gao, Claire Cardie, Sainyam Galhotra• 2026

Related benchmarks

TaskDatasetResultRank
ReasoningARC
Accuracy90.8
269
Multi-task Language UnderstandingMMLU
Top-1 Accuracy74.7
46
Mathematical Problem SolvingMATH
Accuracy (Exact Match)46.2
22
Showing 3 of 3 rows

Other info

Follow for update