# Fairness Testing in Retrieval-Augmented Generation: How Small Perturbations Reveal Bias in Small Language Models

arXiv: https://arxiv.org/abs/2509.26584 · HTML: https://mlsanity.com/paper/2509.26584/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 12 of its 35 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Placed in

- Retrieval-augmented generation / [RAG robustness and adaptive retrieval](https://mlsanity.com/rag/rag-robustness-and-adaptive-retrieval/)
- LLM safety and security / [Membership inference attacks](https://mlsanity.com/safety/membership-inference-attacks/)

## Builds on (most cited first, 12 of 12)

- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval](https://mlsanity.com/paper/2007.00808/) (2020)
- [Bias and Debias in Recommender System: A Survey and Future Directions](https://mlsanity.com/paper/2010.03240/) (2020)
- [Machine Learning Testing: Survey, Landscapes and Horizons](https://mlsanity.com/paper/1906.10742/) (2019)
- [Trustworthiness in Retrieval-Augmented Generation Systems: A Survey](https://mlsanity.com/paper/2409.10102/) (2024)
- [Fairness Testing: A Comprehensive Survey and Analysis of Trends](https://mlsanity.com/paper/2207.10223/) (2022)
- [FairRAG: Fair Human Generation via Fair Retrieval Augmentation](https://mlsanity.com/paper/2403.19964/) (2024)
- [Benchmarking Adversarial Robustness to Bias Elicitation in Large Language Models: Scalable Automated Assessment with LLM-as-a-Judge](https://mlsanity.com/paper/2504.07887/) (2025)
- [MTTM: Metamorphic Testing for Textual Content Moderation Software](https://mlsanity.com/paper/2302.05706/) (2023)
- [METAL: Metamorphic Testing Framework for Analyzing Large-Language Model Qualities](https://mlsanity.com/paper/2312.06056/) (2023)
- [Metamorphic Testing for Fairness Evaluation in Large Language Models: Identifying Intersectional Bias in LLaMA and GPT](https://mlsanity.com/paper/2504.07982/) (2025)
- [Efficient Fairness Testing in Large Language Models: Prioritizing Metamorphic Relations for Bias Detection](https://mlsanity.com/paper/2505.07870/) (2025)

## Built on by (most cited first, 1 of 1)

- [Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinions](https://mlsanity.com/paper/2604.12138/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are widely used across multiple domains but continue to raise concerns regarding security and fairness. Beyond known attack vectors such as data poisoning and prompt injection, LLMs are also vulnerable to fairness bugs. These refer to unintended behaviors influenced by sensitive demographic cues (e.g., race or sexual orientation) that should not affect outcomes. Another key issue is hallucination, where models generate plausible yet false information. Retrieval-Augmented Generation (RAG) has emerged as a strategy to mitigate hallucinations by combining external retrieval with text generation. However, its adoption raises new fairness concerns, as the retrieved content itself may surface or amplify bias. This study conducts fairness testing through metamorphic testing (MT), introducing controlled demographic perturbations in prompts to assess fairness in sentiment analysis performed by three Small Language Models (SLMs) hosted on HuggingFace (Llama-3.2-3B-Instruct, Mistral-7B-Instruct-v0.3, and Llama-3.1-Nemotron-8B), each integrated into a RAG pipeline. Results show that minor demographic variations can break up to one third of metamorphic relations (MRs). A detailed analysis of these failures reveals a consistent bias hierarchy, with perturbations involving racial cues being the predominant cause of the violations. In addition to offering a comparative evaluation, this work reinforces that the retrieval component in RAG must be carefully curated to prevent bias amplification. The findings serve as a practical alert for developers, testers and small organizations aiming to adopt accessible SLMs without compromising fairness or reliability.