@inproceedings{pan-etal-2026-context,
title = "Context Attribution with Multi-Armed Bandit Optimization",
author = "Pan, Deng and
Murugesan, Keerthiram and
Hua, Ting and
Moniz, Nuno and
Chawla, Nitesh V.",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.565/",
pages = "11651--11662",
ISBN = "979-8-89176-395-1",
abstract = "Understanding which parts of the retrieved context contribute to a large language model{'}s generated answer is essential for building interpretable and trustworthy retrieval-augmented generation. We propose a novel framework that formulates context attribution as a combinatorial multi-armed bandit problem. We utilize Linear Thompson Sampling to efficiently identify the most influential context segments while minimizing the number of model queries. Our reward function leverages token log-probabilities to measure how well a subset of segments supports the original response, making it applicable to both open-source and black-box API-based models. Unlike SHAP and other perturbation-based methods that sample subsets uniformly, our approach adaptively prioritizes informative subsets based on posterior estimates of segment relevance, reducing computational costs. Experiments on multiple QA benchmarks demonstrate that our method achieves up to 30{\%} reduction in model queries while matching or exceeding the attribution quality of existing approaches."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pan-etal-2026-context">
<titleInfo>
<title>Context Attribution with Multi-Armed Bandit Optimization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Deng</namePart>
<namePart type="family">Pan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Keerthiram</namePart>
<namePart type="family">Murugesan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ting</namePart>
<namePart type="family">Hua</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nuno</namePart>
<namePart type="family">Moniz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nitesh</namePart>
<namePart type="given">V</namePart>
<namePart type="family">Chawla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Understanding which parts of the retrieved context contribute to a large language model’s generated answer is essential for building interpretable and trustworthy retrieval-augmented generation. We propose a novel framework that formulates context attribution as a combinatorial multi-armed bandit problem. We utilize Linear Thompson Sampling to efficiently identify the most influential context segments while minimizing the number of model queries. Our reward function leverages token log-probabilities to measure how well a subset of segments supports the original response, making it applicable to both open-source and black-box API-based models. Unlike SHAP and other perturbation-based methods that sample subsets uniformly, our approach adaptively prioritizes informative subsets based on posterior estimates of segment relevance, reducing computational costs. Experiments on multiple QA benchmarks demonstrate that our method achieves up to 30% reduction in model queries while matching or exceeding the attribution quality of existing approaches.</abstract>
<identifier type="citekey">pan-etal-2026-context</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.565/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>11651</start>
<end>11662</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Context Attribution with Multi-Armed Bandit Optimization
%A Pan, Deng
%A Murugesan, Keerthiram
%A Hua, Ting
%A Moniz, Nuno
%A Chawla, Nitesh V.
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F pan-etal-2026-context
%X Understanding which parts of the retrieved context contribute to a large language model’s generated answer is essential for building interpretable and trustworthy retrieval-augmented generation. We propose a novel framework that formulates context attribution as a combinatorial multi-armed bandit problem. We utilize Linear Thompson Sampling to efficiently identify the most influential context segments while minimizing the number of model queries. Our reward function leverages token log-probabilities to measure how well a subset of segments supports the original response, making it applicable to both open-source and black-box API-based models. Unlike SHAP and other perturbation-based methods that sample subsets uniformly, our approach adaptively prioritizes informative subsets based on posterior estimates of segment relevance, reducing computational costs. Experiments on multiple QA benchmarks demonstrate that our method achieves up to 30% reduction in model queries while matching or exceeding the attribution quality of existing approaches.
%U https://aclanthology.org/2026.findings-acl.565/
%P 11651-11662
Markdown (Informal)
[Context Attribution with Multi-Armed Bandit Optimization](https://aclanthology.org/2026.findings-acl.565/) (Pan et al., Findings 2026)
ACL
- Deng Pan, Keerthiram Murugesan, Ting Hua, Nuno Moniz, and Nitesh V. Chawla. 2026. Context Attribution with Multi-Armed Bandit Optimization. In Findings of the Association for Computational Linguistics: ACL 2026, pages 11651–11662, San Diego, California, United States. Association for Computational Linguistics.