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

TL;DR

RankFlow introduces a multi-role LLM-based reranking workflow that enhances passage relevance sorting in IR systems by leveraging specialized roles for query understanding, passage summarization, and relevance assessment.

Contribution

This work presents a novel multi-role LLM-driven reranking workflow that significantly improves IR performance on benchmark datasets.

Findings

RankFlow outperforms existing methods on TREC-DL, BEIR, and NovelEval benchmarks.

The multi-role approach effectively interprets queries and assesses passages.

Each role's contribution to overall performance is systematically analyzed.

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.