# No LLM is Free From Bias: A Comprehensive Study of Bias Evaluation in Large Language Models

arXiv: https://arxiv.org/abs/2503.11985 · HTML: https://mlsanity.com/paper/2503.11985/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 15 Mar 2025.

## Builds on (most cited first, 21 of 21)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems](https://mlsanity.com/paper/1805.04508/) (2018)
- ["I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset](https://mlsanity.com/paper/2205.09209/) (2022)
- [UnQovering Stereotyping Biases via Underspecified Questions](https://mlsanity.com/paper/2010.02428/) (2020)
- [Unmasking and Quantifying Racial Bias of Large Language Models in Medical Report Generation](https://mlsanity.com/paper/2401.13867/) (2024)
- [BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs](https://mlsanity.com/paper/2407.10241/) (2024)
- [Beyond Binary Gender Labels: Revealing Gender Biases in LLMs through Gender-Neutral Name Predictions](https://mlsanity.com/paper/2407.05271/) (2024)
- [Quite Good, but Not Enough: Nationality Bias in Large Language Models -- A Case Study of ChatGPT](https://mlsanity.com/paper/2405.06996/) (2024)
- [LLMs with Industrial Lens: Deciphering the Challenges and Prospects -- A Survey](https://mlsanity.com/paper/2402.14558/) (2024)
- [No Size Fits All: The Perils and Pitfalls of Leveraging LLMs Vary with Company Size](https://mlsanity.com/paper/2408.01444/) (2024)

## Built on by (most cited first, 4 of 4)

- [Algorithms Trained on Normal Chest X-rays Can Predict Health Insurance Types](https://mlsanity.com/paper/2511.11030/) (2025)
- [Polar: A Benchmark for Evaluating Political Bias in LLMs](https://mlsanity.com/paper/2606.12922/) (2026)
- [AIAuditTrack: A Framework for AI Security system](https://mlsanity.com/paper/2512.20649/) (2025)
- [Analysis Of Linguistic Stereotypes in Single and Multi-Agent Generative AI Architectures](https://mlsanity.com/paper/2603.18729/) (2026)

## Abstract (arXiv)

Advancements in Large Language Models (LLMs) have increased the performance of different natural language understanding as well as generation tasks. Although LLMs have breached the state-of-the-art performance in various tasks, they often reflect different forms of bias present in the training data. In the light of this perceived limitation, we provide a unified evaluation of benchmarks using a set of representative small and medium-sized LLMs that cover different forms of biases starting from physical characteristics to socio-economic categories. Moreover, we propose five prompting approaches to carry out the bias detection task across different aspects of bias. Further, we formulate three research questions to gain valuable insight in detecting biases in LLMs using different approaches and evaluation metrics across benchmarks. The results indicate that each of the selected LLMs suffer from one or the other form of bias with the Phi-3.5B model being the least biased. Finally, we conclude the paper with the identification of key challenges and possible future directions.