# An Empirical Study of Metrics to Measure Representational Harms in Pre-Trained Language Models

arXiv: https://arxiv.org/abs/2301.09211 · HTML: https://mlsanity.com/paper/2301.09211/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 48 references that are in the map.
- First posted on arXiv 22 Jan 2023.

## Builds on (most cited first, 28 of 28)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [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)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection](https://mlsanity.com/paper/2203.09509/) (2022)
- [BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation](https://mlsanity.com/paper/2101.11718/) (2021)
- [On Measuring Social Biases in Sentence Encoders](https://mlsanity.com/paper/1903.10561/) (2019)
- [Measuring Bias in Contextualized Word Representations](https://mlsanity.com/paper/1906.07337/) (2019)
- [Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems](https://mlsanity.com/paper/1805.04508/) (2018)
- [Latent Hatred: A Benchmark for Understanding Implicit Hate Speech](https://mlsanity.com/paper/2109.05322/) (2021)
- [Gender Bias in Contextualized Word Embeddings](https://mlsanity.com/paper/1904.03310/) (2019)
- [Societal Biases in Language Generation: Progress and Challenges](https://mlsanity.com/paper/2105.04054/) (2021)
- [Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Do](https://mlsanity.com/paper/2103.11790/) (2021)
- [Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases](https://mlsanity.com/paper/2006.03955/) (2020)
- [On the Intrinsic and Extrinsic Fairness Evaluation Metrics for Contextualized Language Representations](https://mlsanity.com/paper/2203.13928/) (2022)
- [Investigating African-American Vernacular English in Transformer-Based Text Generation](https://mlsanity.com/paper/2010.02510/) (2020)
- [Perturbation Augmentation for Fairer NLP](https://mlsanity.com/paper/2205.12586/) (2022)
- [Controlling Style in Generated Dialogue](https://mlsanity.com/paper/2009.10855/) (2020)

## Built on by (most cited first, 24 of 24)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- [Evaluating Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2310.19736/) (2023)
- [ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming](https://mlsanity.com/paper/2404.08676/) (2024)
- [AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies](https://mlsanity.com/paper/2407.17436/) (2024)
- [LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models](https://mlsanity.com/paper/2406.05113/) (2024)
- [BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models](https://mlsanity.com/paper/2408.12798/) (2024)
- [Airavata: Introducing Hindi Instruction-tuned LLM](https://mlsanity.com/paper/2401.15006/) (2024)
- [Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment](https://mlsanity.com/paper/2311.09433/) (2023)
- [The Better Angels of Machine Personality: How Personality Relates to LLM Safety](https://mlsanity.com/paper/2407.12344/) (2024)
- [A Causal Explainable Guardrails for Large Language Models](https://mlsanity.com/paper/2405.04160/) (2024)
- [Beyond Behaviorist Representational Harms: A Plan for Measurement and Mitigation](https://mlsanity.com/paper/2402.01705/) (2024)
- [K2-Think: A Parameter-Efficient Reasoning System](https://mlsanity.com/paper/2509.07604/) (2025)
- [Developing Safe and Responsible Large Language Model : Can We Balance Bias Reduction and Language Understanding in Large Language Models?](https://mlsanity.com/paper/2404.01399/) (2024)
- [LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Inconsistencies](https://mlsanity.com/paper/2412.15035/) (2024)
- [Harm Mitigation in Recommender Systems under User Preference Dynamics](https://mlsanity.com/paper/2406.09882/) (2024)
- [Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectives](https://mlsanity.com/paper/2506.09656/) (2025)
- [MetaAID 2.5: A Secure Framework for Developing Metaverse Applications via Large Language Models](https://mlsanity.com/paper/2312.14480/) (2023)
- [BeanCounter: A low-toxicity, large-scale, and open dataset of business-oriented text](https://mlsanity.com/paper/2409.17827/) (2024)
- [Don't Change My View: Ideological Bias Auditing in Large Language Models](https://mlsanity.com/paper/2509.12652/) (2025)
- [A Multifaceted Analysis of Social Biases in Large Language Models](https://mlsanity.com/paper/2512.15792/) (2025)
- [Compounding Disadvantage: Auditing Intersectional Bias in LLM-Generated Explanations Across Indian and American STEM Education](https://mlsanity.com/paper/2601.14506/) (2026)
- [Navigating the Sea of LLM Evaluation: Investigating Bias in Toxicity Benchmarks](https://mlsanity.com/paper/2605.10639/) (2026)

## Abstract (arXiv)

Large-scale Pre-Trained Language Models (PTLMs) capture knowledge from massive human-written data which contains latent societal biases and toxic contents. In this paper, we leverage the primary task of PTLMs, i.e., language modeling, and propose a new metric to quantify manifested implicit representational harms in PTLMs towards 13 marginalized demographics. Using this metric, we conducted an empirical analysis of 24 widely used PTLMs. Our analysis provides insights into the correlation between the proposed metric in this work and other related metrics for representational harm. We observe that our metric correlates with most of the gender-specific metrics in the literature. Through extensive experiments, we explore the connections between PTLMs architectures and representational harms across two dimensions: depth and width of the networks. We found that prioritizing depth over width, mitigates representational harms in some PTLMs. Our code and data can be found at https://github.com/microsoft/SafeNLP.