# Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Tasks

arXiv: https://arxiv.org/abs/2210.02938 · HTML: https://mlsanity.com/paper/2210.02938/

## Key facts

- Cited by 48 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 58 references that are in the map.
- First posted on arXiv 6 Oct 2022.

## Builds on (most cited first, 30 of 43)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluation](https://mlsanity.com/paper/1708.00055/) (2017)
- [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)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation](https://mlsanity.com/paper/2101.11718/) (2021)
- [HateXplain: A Benchmark Dataset for Explainable Hate Speech Detection](https://mlsanity.com/paper/2012.10289/) (2020)
- [On Measuring Social Biases in Sentence Encoders](https://mlsanity.com/paper/1903.10561/) (2019)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [Measuring Bias in Contextualized Word Representations](https://mlsanity.com/paper/1906.07337/) (2019)
- [Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Them](https://mlsanity.com/paper/1903.03862/) (2019)
- [Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems](https://mlsanity.com/paper/1805.04508/) (2018)
- [Measuring and Reducing Gendered Correlations in Pre-trained Models](https://mlsanity.com/paper/2010.06032/) (2020)
- [Counterfactual Data Augmentation for Mitigating Gender Stereotypes in Languages with Rich Morphology](https://mlsanity.com/paper/1906.04571/) (2019)
- [BERT has a Mouth, and It Must Speak: BERT as a Markov Random Field Language Model](https://mlsanity.com/paper/1902.04094/) (2019)
- [Towards Debiasing Sentence Representations](https://mlsanity.com/paper/2007.08100/) (2020)
- [Learning Gender-Neutral Word Embeddings](https://mlsanity.com/paper/1809.01496/) (2018)
- [Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases](https://mlsanity.com/paper/2006.03955/) (2020)
- [Intrinsic Bias Metrics Do Not Correlate with Application Bias](https://mlsanity.com/paper/2012.15859/) (2020)
- [Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologies](https://mlsanity.com/paper/2108.12084/) (2021)
- [On Measuring and Mitigating Biased Inferences of Word Embeddings](https://mlsanity.com/paper/1908.09369/) (2019)
- [Debiasing Pre-trained Contextualised Embeddings](https://mlsanity.com/paper/2101.09523/) (2021)

## Built on by (most cited first, 30 of 48)

- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Word Embeddings Are Steers for Language Models](https://mlsanity.com/paper/2305.12798/) (2023)
- [Bias in Language Models: Beyond Trick Tests and Toward RUTEd Evaluation](https://mlsanity.com/paper/2402.12649/) (2024)
- [Social Bias Evaluation for Large Language Models Requires Prompt Variations](https://mlsanity.com/paper/2407.03129/) (2024)
- [MABEL: Attenuating Gender Bias using Textual Entailment Data](https://mlsanity.com/paper/2210.14975/) (2022)
- [Dissociating language and thought in large language models](https://mlsanity.com/paper/2301.06627/) (2023)
- [On the Independence of Association Bias and Empirical Fairness in Language Models](https://mlsanity.com/paper/2304.10153/) (2023)
- [Twists, Humps, and Pebbles: Multilingual Speech Recognition Models Exhibit Gender Performance Gaps](https://mlsanity.com/paper/2402.17954/) (2024)
- [BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs](https://mlsanity.com/paper/2407.10241/) (2024)
- [Understanding the Effect of Model Compression on Social Bias in Large Language Models](https://mlsanity.com/paper/2312.05662/) (2023)
- [Evaluating Gender Bias of Pre-trained Language Models in Natural Language Inference by Considering All Labels](https://mlsanity.com/paper/2309.09697/) (2023)
- [On the Intrinsic Self-Correction Capability of LLMs: Uncertainty and Latent Concept](https://mlsanity.com/paper/2406.02378/) (2024)
- [Co2PT: Mitigating Bias in Pre-trained Language Models through Counterfactual Contrastive Prompt Tuning](https://mlsanity.com/paper/2310.12490/) (2023)
- [Safety Alignment via Constrained Knowledge Unlearning](https://mlsanity.com/paper/2505.18588/) (2025)
- [In-Contextual Gender Bias Suppression for Large Language Models](https://mlsanity.com/paper/2309.07251/) (2023)
- [Comparing Intrinsic Gender Bias Evaluation Measures without using Human Annotated Examples](https://mlsanity.com/paper/2301.12074/) (2023)
- [Evaluating the Effect of Retrieval Augmentation on Social Biases](https://mlsanity.com/paper/2502.17611/) (2025)
- [Evaluating Bias and Fairness in Gender-Neutral Pretrained Vision-and-Language Models](https://mlsanity.com/paper/2310.17530/) (2023)
- [On the Origins of Bias in NLP through the Lens of the Jim Code](https://mlsanity.com/paper/2305.09281/) (2023)
- [Solving NLP Problems through Human-System Collaboration: A Discussion-based Approach](https://mlsanity.com/paper/2305.11789/) (2023)
- [The Gaps between Pre-train and Downstream Settings in Bias Evaluation and Debiasing](https://mlsanity.com/paper/2401.08511/) (2024)
- [Smaller Large Language Models Can Do Moral Self-Correction](https://mlsanity.com/paper/2410.23496/) (2024)
- [Investigating Social Bias Changes in Quantized Language Models](https://mlsanity.com/paper/2602.06181/) (2026)
- [A Predictive Factor Analysis of Social Biases and Task-Performance in Pretrained Masked Language Models](https://mlsanity.com/paper/2310.12936/) (2023)
- [Fair Text Classification with Wasserstein Independence](https://mlsanity.com/paper/2311.12689/) (2023)
- [Eagle: Ethical Dataset Given from Real Interactions](https://mlsanity.com/paper/2402.14258/) (2024)
- [Why Don't Prompt-Based Fairness Metrics Correlate?](https://mlsanity.com/paper/2406.05918/) (2024)
- [A Study of Nationality Bias in Names and Perplexity using Off-the-Shelf Affect-related Tweet Classifiers](https://mlsanity.com/paper/2407.01834/) (2024)
- [Exploring Gender Bias in Large Language Models: An In-depth Dive into the German Language](https://mlsanity.com/paper/2507.16557/) (2025)
- [Bias after Prompting: Persistent Discrimination in Large Language Models](https://mlsanity.com/paper/2509.08146/) (2025)

## Abstract (arXiv)

We study the relationship between task-agnostic intrinsic and task-specific extrinsic social bias evaluation measures for Masked Language Models (MLMs), and find that there exists only a weak correlation between these two types of evaluation measures. Moreover, we find that MLMs debiased using different methods still re-learn social biases during fine-tuning on downstream tasks. We identify the social biases in both training instances as well as their assigned labels as reasons for the discrepancy between intrinsic and extrinsic bias evaluation measurements. Overall, our findings highlight the limitations of existing MLM bias evaluation measures and raise concerns on the deployment of MLMs in downstream applications using those measures.