Beyond Relevance: Utility-Centric Retrieval in the LLM Era

TL;DR

This paper discusses shifting from relevance-based to utility-centric retrieval systems in the era of large language models, emphasizing how retrieval should support task accomplishment rather than just relevance.

Contribution

It introduces a unified framework for utility-centric retrieval tailored to LLMs, guiding the design of retrieval systems aligned with LLM information needs.

Findings

Retrieval effectiveness should be evaluated by contribution to generation quality.

A framework distinguishes LLM-agnostic and LLM-specific utility.

Guidance for designing retrieval systems for LLM-based information access.

Abstract

Information retrieval systems have traditionally optimized for topical relevance-the degree to which retrieved documents match a query. However, relevance only approximates a deeper goal: utility, namely, whether retrieved information helps accomplish a user's underlying task. The emergence of retrieval-augmented generation (RAG) fundamentally changes this paradigm. Retrieved documents are no longer consumed directly by users but instead serve as evidence for large language models (LLMs) that produce answers. As a result, retrieval effectiveness must be evaluated by its contribution to generation quality rather than by relevance-based ranking metrics alone. This tutorial argues that retrieval objectives are evolving from relevance-centric optimization toward LLM-centric utility. We present a unified framework covering LLM-agnostic versus LLM-specific utility, context-independent versus context-dependent utility, and the connection with LLM information needs and agentic RAG. By synthesizing recent advances, the tutorial provides conceptual foundations and practical guidance for designing retrieval systems aligned with the requirements of LLM-based information access.