IRCLLGApr 10, 2025

How do Large Language Models Understand Relevance? A Mechanistic Interpretability Perspective

arXiv:2504.07898v15 citationsh-index: 5ACM Transactions on Information Systems
Originality Incremental advance
AI Analysis

This work provides insights into the mechanisms of relevance understanding in LLMs, which could inform future research on applying LLMs to information retrieval, though it is incremental as it explores an underexplored aspect without claiming broad performance gains.

The paper investigates how large language models (LLMs) internally assess relevance for information retrieval tasks, using mechanistic interpretability to identify a multi-stage process involving early-layer information extraction, middle-layer instruction processing, and later-layer judgment generation.

Recent studies have shown that large language models (LLMs) can assess relevance and support information retrieval (IR) tasks such as document ranking and relevance judgment generation. However, the internal mechanisms by which off-the-shelf LLMs understand and operationalize relevance remain largely unexplored. In this paper, we systematically investigate how different LLM modules contribute to relevance judgment through the lens of mechanistic interpretability. Using activation patching techniques, we analyze the roles of various model components and identify a multi-stage, progressive process in generating either pointwise or pairwise relevance judgment. Specifically, LLMs first extract query and document information in the early layers, then process relevance information according to instructions in the middle layers, and finally utilize specific attention heads in the later layers to generate relevance judgments in the required format. Our findings provide insights into the mechanisms underlying relevance assessment in LLMs, offering valuable implications for future research on leveraging LLMs for IR tasks.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes