RankFlow: A Multi-Role Collaborative Reranking Workflow Utilizing Large Language Models

TL;DR

The experimental results demonstrate that RankFlow surpasses state-of-the-art methods on well-established IR benchmarks, including TREC-DL, BEIR, and NovelEval, and achieves an improvement of over 5 points in NDCG@10 than RankGPT-4.

Abstract

In an Information Retrieval (IR) system, reranking plays a critical role by sorting candidate passages according to their relevance to a specific query. This process demands a nuanced understanding of the variations among passages linked to the query. In this work, we introduce RankFlow, a multi-role reranking workflow that leverages the capabilities of Large Language Models (LLMs) and role specializations to improve reranking performance. RankFlow enlists LLMs to fulfill four distinct roles: the query Rewriter, the pseudo Answerer, the passage Summarizer, and the Reranker. This orchestrated approach enables RankFlow to: (1) accurately interpret queries, (2) draw upon LLMs' extensive pre-existing knowledge, (3) distill passages into concise versions, and (4) assess passages in a comprehensive manner, resulting in notably better reranking results. Our experimental results reveal that RankFlow outperforms existing leading approaches on widely recognized IR benchmarks, such as TREC-DL, BEIR, and NovelEval. Additionally, we investigate the individual contributions of each role in RankFlow.

Figures (1)

• Figure 1: Overview of RankFlow . RankFlow is composed of four well-defined expert roles: Rewriter, Answerer, Summarizer, and Reranker, each designed to address specific issues in passage reranking. These roles work sequentially to handle the ranking task.