# RankEvolve: Automating the Discovery of Retrieval Algorithms via LLM-Driven Evolution

Source: [RankEvolve: Automating the Discovery of Retrieval Algorithms via LLM-Driven Evolution](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-28T06:00:02.676Z

Authors: Jinming Nian, Fangchen Li, Dae Hoon Park, Yi Fang

Published: 2026-02-18T22:53:18.000Z

## Abstract

Retrieval algorithms like BM25 and query likelihood with Dirichlet smoothing remain strong and efficient first-stage rankers, yet improvements have mostly relied on parameter tuning and human intuition. We investigate whether a large language model, guided by an evaluator and evolutionary search, can automatically discover improved lexical retrieval algorithms. We introduce RankEvolve, a program evolution setup based on AlphaEvolve, in which candidate ranking algorithms are represented as executable code and iteratively mutated, recombined, and selected based on retrieval performance across 12 IR datasets from BEIR and BRIGHT. RankEvolve starts from two seed programs: BM25 and query likelihood with Dirichlet smoothing. The evolved algorithms are novel, effective, and show promising transfer to the full BEIR and BRIGHT benchmarks as well as TREC DL 19 and 20. Our results suggest that evaluator-guided LLM program evolution is a practical path towards automatic discovery of novel ranking algorithms.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Information Retrieval on BRIGHT unseen 6 subsets (test)](<https://www.sota2.com/research/sota/information-retrieval-on-bright-unseen-6-subsets-test.md>): Information Retrieval; BRIGHT unseen 6 subsets (test); nDCG@10: 11.79.

- [Information Retrieval on BEIR unseen 8 subsets (test)](<https://www.sota2.com/research/sota/information-retrieval-on-beir-unseen-8-subsets-test.md>): Information Retrieval; BEIR unseen 8 subsets (test); nDCG@10: 47.9.

- [Information Retrieval on TREC DL 19 20 (test)](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test.md>): Information Retrieval; TREC DL 19 20 (test); nDCG@10: 64.57.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2602.16932>)

[PDF](<https://arxiv.org/pdf/2602.16932.pdf>)

[Implementation](<https://github.com/fangchenli/ranking-evolved>)