# Active Learners as Efficient PRP Rerankers

arXiv: https://arxiv.org/abs/2605.14236 · HTML: https://mlsanity.com/paper/2605.14236/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 14 May 2026.

## Builds on (most cited first, 12 of 12)

- [Large Language Models for Information Retrieval: A Survey](https://mlsanity.com/paper/2308.07107/) (2023)
- [Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting](https://mlsanity.com/paper/2306.17563/) (2023)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [Simple, Robust and Optimal Ranking from Pairwise Comparisons](https://mlsanity.com/paper/1512.08949/) (2015)
- [Don't Forget to Connect! Improving RAG with Graph-based Reranking](https://mlsanity.com/paper/2405.18414/) (2024)
- [TourRank: Utilizing Large Language Models for Documents Ranking with a Tournament-Inspired Strategy](https://mlsanity.com/paper/2406.11678/) (2024)
- [The Comparative Trap: Pairwise Comparisons Amplifies Biased Preferences of LLM Evaluators](https://mlsanity.com/paper/2406.12319/) (2024)
- [The Sample Complexity of Best-k Items Selection from Pairwise Comparisons](https://mlsanity.com/paper/2007.03133/) (2020)
- [REALM: Recursive Relevance Modeling for LLM-based Document Re-Ranking](https://mlsanity.com/paper/2508.18379/) (2025)
- [Contextual Relevance and Adaptive Sampling for LLM-Based Document Reranking](https://mlsanity.com/paper/2511.01208/) (2025)
- [Are Optimal Algorithms Still Optimal? Rethinking Sorting in LLM-Based Pairwise Ranking with Batching and Caching](https://mlsanity.com/paper/2505.24643/) (2025)
- [LLM Optimization Unlocks Real-Time Pairwise Reranking](https://mlsanity.com/paper/2511.07555/) (2025)

## Abstract (arXiv)

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.