# Evaluating Gender Bias of Pre-trained Language Models in Natural Language Inference by Considering All Labels

arXiv: https://arxiv.org/abs/2309.09697 · HTML: https://mlsanity.com/paper/2309.09697/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 37 references that are in the map.
- First posted on arXiv 18 Sep 2023.

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (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)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [Measuring and Reducing Gendered Correlations in Pre-trained Models](https://mlsanity.com/paper/2010.06032/) (2020)
- [Intrinsic Bias Metrics Do Not Correlate with Application Bias](https://mlsanity.com/paper/2012.15859/) (2020)
- [On Measuring and Mitigating Biased Inferences of Word Embeddings](https://mlsanity.com/paper/1908.09369/) (2019)
- [OCNLI: Original Chinese Natural Language Inference](https://mlsanity.com/paper/2010.05444/) (2020)
- [On Measures of Biases and Harms in NLP](https://mlsanity.com/paper/2108.03362/) (2021)
- [On the Intrinsic and Extrinsic Fairness Evaluation Metrics for Contextualized Language Representations](https://mlsanity.com/paper/2203.13928/) (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)
- [Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Tasks](https://mlsanity.com/paper/2210.02938/) (2022)
- [Socially Aware Bias Measurements for Hindi Language Representations](https://mlsanity.com/paper/2110.07871/) (2021)
- [Evaluating Gender Bias in Natural Language Inference](https://mlsanity.com/paper/2105.05541/) (2021)
- [Comparing Intrinsic Gender Bias Evaluation Measures without using Human Annotated Examples](https://mlsanity.com/paper/2301.12074/) (2023)
- [Gender Bias Hidden Behind Chinese Word Embeddings: The Case of Chinese Adjectives](https://mlsanity.com/paper/2106.00181/) (2021)

## Built on by (most cited first, 16 of 16)

- [Evaluating Gender Bias in Large Language Models via Chain-of-Thought Prompting](https://mlsanity.com/paper/2401.15585/) (2024)
- [Social Bias Evaluation for Large Language Models Requires Prompt Variations](https://mlsanity.com/paper/2407.03129/) (2024)
- [Sampling-based Pseudo-Likelihood for Membership Inference Attacks](https://mlsanity.com/paper/2404.11262/) (2024)
- [Likelihood-based Mitigation of Evaluation Bias in Large Language Models](https://mlsanity.com/paper/2402.15987/) (2024)
- [Controlled Generation with Prompt Insertion for Natural Language Explanations in Grammatical Error Correction](https://mlsanity.com/paper/2309.11439/) (2023)
- [JBBQ: Japanese Bias Benchmark for Analyzing Social Biases in Large Language Models](https://mlsanity.com/paper/2406.02050/) (2024)
- [BharatBBQ: A Multilingual Bias Benchmark for Question Answering in the Indian Context](https://mlsanity.com/paper/2508.07090/) (2025)
- [The Gaps between Pre-train and Downstream Settings in Bias Evaluation and Debiasing](https://mlsanity.com/paper/2401.08511/) (2024)
- [Hire Me or Not? Examining Language Model's Behavior with Occupation Attributes](https://mlsanity.com/paper/2405.06687/) (2024)
- [Intent-Aware Self-Correction for Mitigating Social Biases in Large Language Models](https://mlsanity.com/paper/2503.06011/) (2025)
- [Eagle: Ethical Dataset Given from Real Interactions](https://mlsanity.com/paper/2402.14258/) (2024)
- [Analyzing Continuous Semantic Shifts with Diachronic Word Similarity Matrices](https://mlsanity.com/paper/2501.09538/) (2025)
- [Analyzing the Safety of Japanese Large Language Models in Stereotype-Triggering Prompts](https://mlsanity.com/paper/2503.01947/) (2025)
- [JUBAKU: An Adversarial Benchmark for Exposing Culturally Grounded Stereotypes in Japanese LLMs](https://mlsanity.com/paper/2603.20581/) (2026)
- [Balanced Multi-Factor In-Context Learning for Multilingual Large Language Models](https://mlsanity.com/paper/2502.11495/) (2025)
- [Large Language Models Are Still Misled by Simple Bias Ensembles](https://mlsanity.com/paper/2505.16522/) (2025)

## Abstract (arXiv)

Discriminatory gender biases have been found in Pre-trained Language Models (PLMs) for multiple languages. In Natural Language Inference (NLI), existing bias evaluation methods have focused on the prediction results of one specific label out of three labels, such as neutral. However, such evaluation methods can be inaccurate since unique biased inferences are associated with unique prediction labels. Addressing this limitation, we propose a bias evaluation method for PLMs, called NLI-CoAL, which considers all the three labels of NLI task. First, we create three evaluation data groups that represent different types of biases. Then, we define a bias measure based on the corresponding label output of each data group. In the experiments, we introduce a meta-evaluation technique for NLI bias measures and use it to confirm that our bias measure can distinguish biased, incorrect inferences from non-biased incorrect inferences better than the baseline, resulting in a more accurate bias evaluation. We create the datasets in English, Japanese, and Chinese, and successfully validate the compatibility of our bias measure across multiple languages. Lastly, we observe the bias tendencies in PLMs of different languages. To our knowledge, we are the first to construct evaluation datasets and measure PLMs' bias from NLI in Japanese and Chinese.