<div class="csl-bib-body">
<div class="csl-entry">Nematov, I., Kalai, T., Kuzmenko, E., Fugagnoli, G., Sacharidis, D., Hose, K., & Sagi, T. (2025). <i>Source Attribution in Retrieval-Augmented Generation</i>. arXiv. https://doi.org/10.48550/ARXIV.2507.04480</div>
</div>
-
dc.identifier.uri
http://hdl.handle.net/20.500.12708/222502
-
dc.description.abstract
While attribution methods, such as Shapley values, are widely used to explain the importance of features or training data in traditional machine learning, their application to Large Language Models (LLMs), particularly within Retrieval-Augmented Generation (RAG) systems, is nascent and challenging. The primary obstacle is the substantial computational cost, where each utility function evaluation involves an expensive LLM call, resulting in direct monetary and time expenses. This paper investigates the feasibility and effectiveness of adapting Shapley-based attribution to identify influential retrieved documents in RAG. We compare Shapley with more computationally tractable approximations and some existing attribution methods for LLM. Our work aims to: (1) systematically apply established attribution principles to the RAG document-level setting; (2) quantify how well SHAP approximations can mirror exact attributions while minimizing costly LLM interactions; and (3) evaluate their practical explainability in identifying critical documents, especially under complex inter-document relationships such as redundancy, complementarity, and synergy. This study seeks to bridge the gap between powerful attribution techniques and the practical constraints of LLM-based RAG systems, offering insights into achieving reliable and affordable RAG explainability.
en
dc.language.iso
en
-
dc.subject
RAG
-
dc.subject
retrieval-augmented generation
-
dc.subject
LLM
-
dc.title
Source Attribution in Retrieval-Augmented Generation
-
dc.type
Preprint
en
dc.type
Preprint
de
dc.identifier.arxiv
2507.04480
-
dc.contributor.affiliation
Université Libre de Bruxelles, Belgium
-
dc.contributor.affiliation
Université Libre de Bruxelles, Belgium
-
dc.contributor.affiliation
Université Libre de Bruxelles, Belgium
-
dc.contributor.affiliation
Université Libre de Bruxelles, Belgium
-
dc.contributor.affiliation
Université Libre de Bruxelles, Belgium
-
dc.contributor.affiliation
Aalborg University, Denmark
-
tuw.researchTopic.id
I1
-
tuw.researchTopic.id
I4
-
tuw.researchTopic.name
Logic and Computation
-
tuw.researchTopic.name
Information Systems Engineering
-
tuw.researchTopic.value
20
-
tuw.researchTopic.value
80
-
tuw.publication.orgunit
E192-02 - Forschungsbereich Databases and Artificial Intelligence
-
tuw.publication.orgunit
E056-23 - Fachbereich Innovative Combinations and Applications of AI and ML (iCAIML)
-
tuw.publisher.doi
10.48550/ARXIV.2507.04480
-
tuw.author.orcid
0009-0005-1794-0669
-
tuw.author.orcid
0000-0001-5022-1483
-
tuw.author.orcid
0000-0002-8916-0128
-
tuw.publisher.server
arXiv
-
wb.sciencebranch
Informatik
-
wb.sciencebranch
Mathematik
-
wb.sciencebranch.oefos
1020
-
wb.sciencebranch.oefos
1010
-
wb.sciencebranch.value
80
-
wb.sciencebranch.value
20
-
item.grantfulltext
none
-
item.cerifentitytype
Publications
-
item.fulltext
no Fulltext
-
item.openairecristype
http://purl.org/coar/resource_type/c_816b
-
item.openairetype
preprint
-
item.languageiso639-1
en
-
crisitem.author.dept
Université Libre de Bruxelles, Belgium
-
crisitem.author.dept
Université Libre de Bruxelles, Belgium
-
crisitem.author.dept
Université Libre de Bruxelles, Belgium
-
crisitem.author.dept
Université Libre de Bruxelles, Belgium
-
crisitem.author.dept
E194-04 - Forschungsbereich Data Science
-
crisitem.author.dept
E192-02 - Forschungsbereich Databases and Artificial Intelligence
-
crisitem.author.dept
Aalborg University, Denmark
-
crisitem.author.orcid
0009-0005-1794-0669
-
crisitem.author.orcid
0000-0002-9803-5193
-
crisitem.author.orcid
0000-0001-5022-1483
-
crisitem.author.orcid
0000-0001-7025-8099
-
crisitem.author.orcid
0000-0002-8916-0128
-
crisitem.author.parentorg
E194 - Institut für Information Systems Engineering