On Listwise Reranking for Corpus Feedback

2025/10/01 by Yoon, Soyoung, Kim, Jongho, Kwon, Daeyong +2
#FOS: Computer and information sciences #Information Retrieval (cs.IR)

paper · doi:10.48550/arxiv.2510.00887

Abstract

Reranker improves retrieval performance by capturing document interactions. At one extreme, graph-aware adaptive retrieval (GAR) represents an information-rich regime, requiring a pre-computed document similarity graph in reranking. However, as such graphs are often unavailable, or incur quadratic memory costs even when available, graph-free rerankers leverage large language model (LLM) calls to achieve competitive performance. We introduce L2G, a novel framework that implicitly induces document graphs from listwise reranker logs. By converting reranker signals into a graph structure, L2G enables scalable graph-based retrieval without the overhead of explicit graph computation. Results on the TREC-DL and BEIR subset show that L2G matches the effectiveness of oracle-based graph methods, while incurring zero additional LLM calls.

Citations

• Test-time Corpus Feedback: From Retrieval to RAG
• REARANK: Reasoning Re-ranking Agent via Reinforcement Learning
• CoRanking: Collaborative Ranking with Small and Large Ranking Agents
• Guiding Retrieval using LLM-based Listwise Rankers
• DemoRank: Selecting Effective Demonstrations for Large Language Models in Ranking Task
• TourRank: Utilizing Large Language Models for Documents Ranking with a Tournament-Inspired Strategy
• RaFe: Ranking Feedback Improves Query Rewriting for RAG
• RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!
• Unsupervised Dense Information Retrieval with Contrastive Learning
• BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models
• Distilling Dense Representations for Ranking using Tightly-Coupled Teachers
• Probabilistic selection of inducing points in sparse Gaussian processes
• Overview of the TREC 2019 deep learning track
• Cumulated gain-based evaluation of IR techniques

Related

• Guiding Retrieval using LLM-based Listwise Rankers
• Zero-Shot Listwise Document Reranking with a Large Language Model
• FIRST: Faster Improved Listwise Reranking with Single Token Decoding
• Beyond Sequential Reranking: Reranker-Guided Search Improves Reasoning Intensive Retrieval
• Don't Forget to Connect! Improving RAG with Graph-based Reranking