mlsanityData to 7 Oct 2026

Evaluating the Effect of Retrieval Augmentation on Social Biases

arXiv 2502.17611 ↗First posted 24 Feb 2025cs.CLcited by 7

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

Abstract (arXiv)

Retrieval Augmented Generation (RAG) has gained popularity as a method for conveniently incorporating novel facts that were not seen during the pre-training stage in Large Language Model (LLM)-based Natural Language Generation (NLG) systems. However, LLMs are known to encode significant levels of unfair social biases. The modulation of these biases by RAG in NLG systems is not well understood. In this paper, we systematically study the relationship between the different components of a RAG system and the social biases presented in the text generated across three languages (i.e. English, Japanese and Chinese) and four social bias types (i.e. gender, race, age and religion). Specifically, using the Bias Question Answering (BBQ) benchmark datasets, we evaluate the social biases in RAG responses from document collections with varying levels of stereotypical biases, employing multiple LLMs used as generators. We find that the biases in document collections are often amplified in the generated responses, even when the generating LLM exhibits a low-level of bias. Our findings raise concerns about the use of RAG as a technique for injecting novel facts into NLG systems and call for careful evaluation of potential social biases in RAG applications before their real-world deployment.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 49 of its 74 references that are in the map. First posted on arXiv 24 Feb 2025.

Builds on

49 of 74 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024No Free Lunch: Retrieval-Augmented Generation Undermines Fairness in LLMs, Even for Vigilant Users10 Oct
  2. 2024Does RAG Introduce Unfairness in LLMs? Evaluating Fairness in Retrieval-Augmented Generation Systems29 Sep
  3. 2024Fact, Fetch, and Reason: A Unified Evaluation of Retrieval-Augmented Generation19 Sep
  4. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection17 Oct

Other references · 45

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,168
  6. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,411
  7. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  8. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
  9. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  10. 2024The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scalecited by 1,104
  11. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  12. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 911
33 more
  1. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  2. 2016Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphscited by 853
  3. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
  4. 2020CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Modelscited by 703
  5. 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 691
  6. 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 683
  7. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
  8. 2019Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Settingcited by 346
  9. 2023RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Modelscited by 272
  10. 2020Measuring and Reducing Gendered Correlations in Pre-trained Modelscited by 222
  11. 2024RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMscited by 200
  12. 2023Continual Pre-training of Language Modelscited by 194
  13. 2021RedditBias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language Modelscited by 149
  14. 2020Intrinsic Bias Metrics Do Not Correlate with Application Biascited by 148
  15. 2019Assessing Social and Intersectional Biases in Contextualized Word Representationscited by 134
  16. 2023Can Knowledge Graphs Reduce Hallucinations in LLMs? : A Surveycited by 92
  17. 2024Summary of a Haystack: A Challenge to Long-Context LLMs and RAG Systemscited by 87
  18. 2022On the Intrinsic and Extrinsic Fairness Evaluation Metrics for Contextualized Language Representationscited by 75
  19. 2023KoBBQ: Korean Bias Benchmark for Question Answeringcited by 71
  20. 2021Unmasking the Mask -- Evaluating Social Biases in Masked Language Modelscited by 60
  21. 2022Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Taskscited by 48
  22. 2024On Large Language Models' Hallucination with Regard to Known Factscited by 43
  23. 2024Modular RAG: Transforming RAG Systems into LEGO-like Reconfigurable Frameworkscited by 41
  24. 2024Decoding Biases: Automated Methods and LLM Judges for Gender Bias Detection in Language Modelscited by 33
  25. 2020Do Neural Ranking Models Intensify Gender Bias?cited by 28
  26. 2023CHBias: Bias Evaluation and Mitigation of Chinese Conversational Language Modelscited by 25
  27. 2024GenderAlign: An Alignment Dataset for Mitigating Gender Bias in Large Language Modelscited by 18
  28. 2023SimCSE++: Improving Contrastive Learning for Sentence Embeddings from Two Perspectivescited by 15
  29. 2023''Fifty Shades of Bias'': Normative Ratings of Gender Bias in GPT Generated English Textcited by 15
  30. 2023Women Are Beautiful, Men Are Leaders: Gender Stereotypes in Machine Translation and Language Modelingcited by 12
  31. 2024Observations on Building RAG Systems for Technical Documentscited by 8
  32. 2023Overview of the TREC 2022 Fair Ranking Trackcited by 7
  33. 2024Evaluating Short-Term Temporal Fluctuations of Social Biases in Social Media Data and Masked Language Modelscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 6.

7 papers.

  1. 2025Investigating Thinking Behaviours of Reasoning-Based Language Models for Social Bias Mitigationcited by 2
  2. 2026Evaluating Social Bias in RAG Systems: When External Context Helps and Reasoning Hurtscited by 1
  3. 2026Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinionscited by 1
  4. 2026Fairness-Aware Retrieval Optimization for Retrieval-Augmented Generationcited by 0
  5. 2026Adversarial Feeds Steer LLM Agent Decisions Against Their Defaultscited by 0
  6. 2026Exploratory As-Analyzed No-Detection of Culturally-Marked Predicate-Triggered PII Amplification in a Synthetic-English RAG Probe: A Predicate-Resource-Confounded Auditcited by 0
  7. 2026Retrieval Sensitivity to Identity Signals in Queriescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024LLMs are Biased Evaluators But Not Biased for Retrieval Augmented Generation
  2. 2025Mitigating Bias in RAG: Controlling the Embedder
  3. 2025Out of Style: RAG's Fragility to Linguistic Variation
  4. 2026Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias
  5. 2025The Other Side of the Coin: Exploring Fairness in Retrieval-Augmented Generation
  6. 2025Retrieval Augmented Generation Evaluation in the Era of Large Language Models: A Comprehensive Survey
  7. 2024BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs
  8. 2024A Comprehensive Survey of Retrieval-Augmented Generation (RAG): Evolution, Current Landscape and Future Directions