@inproceedings{liu-etal-2025-pointwise,
title = "Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation",
author = "Liu, Tianyu and
Qi, Jirui and
He, Paul and
Bisazza, Arianna and
Sachan, Mrinmaya and
Cotterell, Ryan",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.78/",
pages = "1628--1647",
ISBN = "979-8-89176-189-6",
abstract = "Recent work suggests that large language models enhanced with retrieval-augmented generation are easily influenced by the order in which the retrieved documents are presented to the model when solving tasks such as question answering (QA).However, there is no method to date that exploits this phenomenon to improve generation.To fill this gap, in this study, we show that the pointwise mutual information between a context and a question is an effective gauge for language model performance.Importantly, this gauge does not depend on knowing the answer to the question \textit{a priori}.Through experiments on two question-answering datasets using a variety of large language models, we find evidence for an empirical correlation between answer accuracy and pointwise mutual information.Additionally, we propose two methods that use the pointwise mutual information between a document and a question as a gauge for selecting and constructing prompts that lead to better performance, whose effectiveness we demonstrate through experimentation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2025-pointwise">
<titleInfo>
<title>Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tianyu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jirui</namePart>
<namePart type="family">Qi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arianna</namePart>
<namePart type="family">Bisazza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mrinmaya</namePart>
<namePart type="family">Sachan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="family">Cotterell</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>Recent work suggests that large language models enhanced with retrieval-augmented generation are easily influenced by the order in which the retrieved documents are presented to the model when solving tasks such as question answering (QA).However, there is no method to date that exploits this phenomenon to improve generation.To fill this gap, in this study, we show that the pointwise mutual information between a context and a question is an effective gauge for language model performance.Importantly, this gauge does not depend on knowing the answer to the question a priori.Through experiments on two question-answering datasets using a variety of large language models, we find evidence for an empirical correlation between answer accuracy and pointwise mutual information.Additionally, we propose two methods that use the pointwise mutual information between a document and a question as a gauge for selecting and constructing prompts that lead to better performance, whose effectiveness we demonstrate through experimentation.</abstract>
<identifier type="citekey">liu-etal-2025-pointwise</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.78/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>1628</start>
<end>1647</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation
%A Liu, Tianyu
%A Qi, Jirui
%A He, Paul
%A Bisazza, Arianna
%A Sachan, Mrinmaya
%A Cotterell, Ryan
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F liu-etal-2025-pointwise
%X Recent work suggests that large language models enhanced with retrieval-augmented generation are easily influenced by the order in which the retrieved documents are presented to the model when solving tasks such as question answering (QA).However, there is no method to date that exploits this phenomenon to improve generation.To fill this gap, in this study, we show that the pointwise mutual information between a context and a question is an effective gauge for language model performance.Importantly, this gauge does not depend on knowing the answer to the question a priori.Through experiments on two question-answering datasets using a variety of large language models, we find evidence for an empirical correlation between answer accuracy and pointwise mutual information.Additionally, we propose two methods that use the pointwise mutual information between a document and a question as a gauge for selecting and constructing prompts that lead to better performance, whose effectiveness we demonstrate through experimentation.
%U https://aclanthology.org/2025.naacl-long.78/
%P 1628-1647
Markdown (Informal)
[Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation](https://aclanthology.org/2025.naacl-long.78/) (Liu et al., NAACL 2025)
ACL
- Tianyu Liu, Jirui Qi, Paul He, Arianna Bisazza, Mrinmaya Sachan, and Ryan Cotterell. 2025. Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 1628–1647, Albuquerque, New Mexico. Association for Computational Linguistics.