# On Measuring and Mitigating Biased Inferences of Word Embeddings

arXiv: https://arxiv.org/abs/1908.09369 · HTML: https://mlsanity.com/paper/1908.09369/

## Key facts

- Cited by 112 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Aug 2019.

## Built on by (most cited first, 30 of 112)

- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- [Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs](https://mlsanity.com/paper/2311.04892/) (2023)
- [RedditBias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language Models](https://mlsanity.com/paper/2106.03521/) (2021)
- [Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologies](https://mlsanity.com/paper/2108.12084/) (2021)
- [Debiasing Pre-trained Contextualised Embeddings](https://mlsanity.com/paper/2101.09523/) (2021)
- [Sustainable Modular Debiasing of Language Models](https://mlsanity.com/paper/2109.03646/) (2021)
- [A Survey on Fairness in Large Language Models](https://mlsanity.com/paper/2308.10149/) (2023)
- [A Survey on Gender Bias in Natural Language Processing](https://mlsanity.com/paper/2112.14168/) (2021)
- [On Measures of Biases and Harms in NLP](https://mlsanity.com/paper/2108.03362/) (2021)
- [Bias in Large Language Models: Origin, Evaluation, and Mitigation](https://mlsanity.com/paper/2411.10915/) (2024)
- [SeeGULL: A Stereotype Benchmark with Broad Geo-Cultural Coverage Leveraging Generative Models](https://mlsanity.com/paper/2305.11840/) (2023)
- [Theories of "Gender" in NLP Bias Research](https://mlsanity.com/paper/2205.02526/) (2022)
- [Unmasking the Mask -- Evaluating Social Biases in Masked Language Models](https://mlsanity.com/paper/2104.07496/) (2021)
- [Gender Bias in Masked Language Models for Multiple Languages](https://mlsanity.com/paper/2205.00551/) (2022)
- [Fairness in Large Language Models: A Taxonomic Survey](https://mlsanity.com/paper/2404.01349/) (2024)
- [AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq Model](https://mlsanity.com/paper/2208.01448/) (2022)
- [Re-contextualizing Fairness in NLP: The Case of India](https://mlsanity.com/paper/2209.12226/) (2022)
- [Nationality Bias in Text Generation](https://mlsanity.com/paper/2302.02463/) (2023)
- [Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Tasks](https://mlsanity.com/paper/2210.02938/) (2022)
- [BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation](https://mlsanity.com/paper/2210.07626/) (2022)
- [Welcome to the Modern World of Pronouns: Identity-Inclusive Natural Language Processing beyond Gender](https://mlsanity.com/paper/2202.11923/) (2022)
- [Ethical-Advice Taker: Do Language Models Understand Natural Language Interventions?](https://mlsanity.com/paper/2106.01465/) (2021)
- [Quantifying Social Biases Using Templates is Unreliable](https://mlsanity.com/paper/2210.04337/) (2022)
- [The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks](https://mlsanity.com/paper/2210.10040/) (2022)
- [MABEL: Attenuating Gender Bias using Textual Entailment Data](https://mlsanity.com/paper/2210.14975/) (2022)

## Abstract (arXiv)

Word embeddings carry stereotypical connotations from the text they are trained on, which can lead to invalid inferences in downstream models that rely on them. We use this observation to design a mechanism for measuring stereotypes using the task of natural language inference. We demonstrate a reduction in invalid inferences via bias mitigation strategies on static word embeddings (GloVe). Further, we show that for gender bias, these techniques extend to contextualized embeddings when applied selectively only to the static components of contextualized embeddings (ELMo, BERT).