An Iterative Utility Judgment Framework Inspired by Philosophical Relevance via LLMs

TL;DR

This paper introduces ITEM, an iterative framework inspired by philosophical relevance, to improve utility judgments, relevance ranking, and answer generation in retrieval-augmented generation systems using LLMs.

Contribution

The paper proposes a novel iterative utility judgment framework that enhances each component of RAG by aligning with philosophical relevance, improving overall system performance.

Findings

ITEM improves utility judgments over baselines.

ITEM enhances relevance ranking accuracy.

ITEM leads to better answer generation quality.

Abstract

Relevance and utility are two frequently used measures to evaluate the effectiveness of an information retrieval (IR) system. Relevance emphasizes the aboutness of a result to a query, while utility refers to the result's usefulness or value to an information seeker. In retrieval-augmented generation (RAG), high-utility results should be prioritized to feed to LLMs due to their limited input bandwidth. Re-examining RAG's three core components-relevance ranking derived from retrieval models, utility judgments, and answer generation-aligns with Schutz's philosophical system of relevances, which encompasses three types of relevance representing different levels of human cognition that enhance each other. These three RAG components also reflect three cognitive levels for LLMs in question-answering. Therefore, we propose an Iterative utiliTy judgmEnt fraMework (ITEM) to promote each step in RAG. We conducted extensive experiments on retrieval (TREC DL, WebAP), utility judgment task (GTI-NQ), and factoid question-answering (NQ) datasets. Experimental results demonstrate improvements of ITEM in utility judgments, ranking, and answer generation upon representative baselines.