ARXIV:2603.24218 · LLM FAIRNESS · SUBMITTED 02 APR · 02:30 UTC · FRESHNESS STALE

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias

Mahdi Dehghan · Graham McDonald · arXiv

This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

Ship in 2-4 weeks›Score7.0Evidence unverified

Opportunity summary

Pain This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

Evidence 0 refs | 0 sources | 17% coverage

Blocker Evidence unverified

Open Build Read PDF Signal Canvas Track

PROBLEM

This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications. However, relatively little work has investigated the impact of RAG in terms of fairness.

METHOD

Full abstract

Large Language Models (LLMs) enhanced with Retrieval-Augmented Generation (RAG) have achieved substantial improvements in accuracy by grounding their responses in external documents that are relevant to the user's query. However, relatively little work has investigated the impact of RAG in terms of fairness. Particularly, it is not yet known if queries that are associated with certain groups within a fairness category systematically receive higher accuracy, or accuracy improvements in RAG systems compared to LLM-only, a phenomenon we refer to as query group fairness. In this work, we conduct extensive experiments to investigate the impact of three key factors on query group fairness in RAG, namely: Group exposure, i.e., the proportion of documents from each group appearing in the retrieved set, determined by the retriever; Group utility, i.e., the degree to which documents from each group contribute to improving answer accuracy, capturing retriever-generator interactions; and Group attribution, i.e., the extent to which the generator relies on documents from each group when producing responses. We examine group-level average accuracy and accuracy improvements disparities across four fairness categories using three datasets derived from the TREC 2022 Fair Ranking Track for two tasks: article generation and title generation. Our findings show that RAG systems suffer from the query group fairness problem and amplify disparities in terms of average accuracy across queries from different groups, compared to an LLM-only setting. Moreover, group utility, exposure, and attribution can exhibit strong positive or negative correlations with average accuracy or accuracy improvements of queries from that group, highlighting their important role in fair RAG. Our data and code are publicly available from Github.

RESULT

ScienceToStartup currently rates this 7.0/10 on the public viability pass. Our findings show that RAG systems suffer from the query group fairness problem and amplify disparities in terms of average accuracy across queries from…

WHY NOW

LLM Fairness moved forward this cycle; last verified April 2026. Public score 7.0/10. Production flags indicate code availability.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score7.0

PainThis research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

Evidence0 refs | 0 sources | 17% coverage

Blockerno shell-level blocker reported

Analysis summary

This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

Competitive landscape

This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

Segment

LLM Fairness

Adoption evidence

No public code link in the paper record yet

Commercial read

7.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

{ "contract_version": "paper-r2", "paper_id": "bb204ce8-8718-4d23-b7e2-8e1b838d1ad1", "arxiv_id": "2603.24218", "canonical_route": "/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "active_tab": "synced from current hash by the drawer client", "selected_artifact": "who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "endpoints": { "paper_pack": "/api/v1/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias/paper-pack", "build_passport": "/api/v1/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias/build-passport", "mcp_resource": "sciencetostartup://surfaces/paper-workspace" } }

{ "surface": "paper", "mode": "paper", "query": "Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias", "normalized_query": "2603.24218", "route": "/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "paper_ref": "who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "topic_slug": null, "benchmark_ref": null, "dataset_ref": null }

{ "@context": "https://schema.org", "@graph": [ { "@type": "WebPage", "@id": "https://sciencetostartup.com/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias#webpage", "url": "https://sciencetostartup.com/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "name": "Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias", "description": "This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.", "isPartOf": { "@id": "https://sciencetostartup.com/#website" } }, { "@type": "ScholarlyArticle", "@id": "https://sciencetostartup.com/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias#scholarlyArticle", "headline": "Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias", "description": "This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.", "url": "https://sciencetostartup.com/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias", "sameAs": "https://arxiv.org/abs/2603.24218", "identifier": { "@type": "PropertyValue", "propertyID": "arXiv", "value": "2603.24218" }, "isAccessibleForFree": true, "isPartOf": { "@id": "https://sciencetostartup.com/#website" }, "datePublished": "2026-03-25T11:45:52.000Z", "author": [ { "@type": "Person", "name": "Mahdi Dehghan" }, { "@type": "Person", "name": "Graham McDonald" } ], "additionalProperty": [ { "@type": "PropertyValue", "propertyID": "viabilityScore", "value": 7 }, { "@type": "PropertyValue", "propertyID": "researchDomain", "value": "LLM Fairness" }, { "@type": "PropertyValue", "propertyID": "commercialReadiness", "value": "code" } ] }, { "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://sciencetostartup.com" }, { "@type": "ListItem", "position": 2, "name": "LLM Fairness", "item": "https://sciencetostartup.com/topics" }, { "@type": "ListItem", "position": 3, "name": "Who Benefits from RAG? The Role of Exposure, Utility and Att", "item": "https://sciencetostartup.com/paper/who-benefits-from-rag-the-role-of-exposure-utility-and-attribution-bias" } ] } ] }

Competitive landscape

This research quantifies and addresses fairness disparities in Retrieval-Augmented Generation (RAG) systems, offering a path to more equitable LLM applications.

Segment

LLM Fairness

Adoption evidence

No public code link in the paper record yet

Commercial read

7.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias

Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline