Pairwise Ranking Prompting (PRP) elicits pairwise preference judgments from an LLM, which are then aggregated into a ranking, usually via classical sorting algorithms. However, judgments are noisy, order-sensitive, and sometimes intransitive, so sorting assumptions do not match the setting. Because sorting aims to recover a full permutation, truncating it to meet a call budget does not produce a dependable top-K. We thus reframe PRP reranking as active learning from noisy pairwise comparisons and show that active rankers are drop-in replacements that improve NDCG@10 per call in the call-constrained regime. Our noise-robust framework also introduces a randomized-direction oracle that uses a single LLM call per pair. This approach converts systematic position bias into zero-mean noise, enabling unbiased aggregate ranking without the cost of bidirectional calls.
@article{arxiv.2605.14236,
title = {Active Learners as Efficient PRP Rerankers},
author = {Jeremías Figueiredo Paschmann and Juan Kaplan and Francisco Nattero and Santiago Barron and Juan Wisznia and Luciano del Corro},
journal= {arXiv preprint arXiv:2605.14236},
year = {2026}
}