Gender stereotypes are manifest in most of the world's languages and are consequently propagated or amplified by NLP systems. Although research has focused on mitigating gender stereotypes in English, the approaches that are commonly employed produce ungrammatical sentences in morphologically rich languages. We present a novel approach for converting between masculine-inflected and feminine-inflected sentences in such languages. For Spanish and Hebrew, our approach achieves F1 scores of 82% and 73% at the level of tags and accuracies of 90% and 87% at the level of forms. By evaluating our approach using four different languages, we show that, on average, it reduces gender stereotyping by a factor of 2.5 without any sacrifice to grammaticality.
In this work, we investigate the presence of occupational gender stereotypes in sentiment analysis models. Such a task has implications for reducing implicit biases in these models, which are being applied to an increasingly wide variety of downstream tasks. We release a new gender-balanced dataset of 800 sentences pertaining to specific professions and propose a methodology for using it as a test bench to evaluate sentiment analysis models. We evaluate the presence of occupational gender stereotypes in 3 different models using our approach, and explore their relationship with societal percept…
Revisiting The Classics: A Study on Identifying and Rectifying Gender Stereotypes in Rhymes and Poems
Sri Sathya Sai Institute of Higher Learning（斯里·萨蒂亚·赛高等学习学院）；GESIS - Leibniz Institute for the Social Sciences（莱布尼茨社会科学研究所（GESIS））；Citridot Solutions（西特里多特解决方案公司）；University of Tartu（塔尔图大学）；IIT Ropar（罗帕尔印度理工学院）
School of Journalism and Communication, Nanjing University（南京大学新闻传播学院）；School of Systems Science, Beijing Normal University（北京师范大学系统科学学院）；School of Computing and Information, University of Pittsburgh（匹兹堡大学计算与信息学院）
Our analysis of thousands of movies and books reveals how these cultural products weave stereotypical gender roles into morality tales and perpetuate gender inequality through storytelling. Using the word embedding techniques, we reveal the constructed emotional dependency of female characters on male characters in stories.
Manipal Institute of Technology（马尼帕尔理工学院）；MAHE（马尼帕尔高等教育学院）；University of Texas, Austin（德克萨斯大学奥斯汀分校）
As the use of natural language processing increases in our day-to-day life, the need to address gender bias inherent in these systems also amplifies. This is because the inherent bias interferes with the semantic structure of the output of these systems while performing tasks like machine translation. While research is being done in English to quantify and mitigate bias, debiasing methods in Indic Languages are either relatively nascent or absent for some Indic languages altogether. Most Indic languages are gendered, i.e., each noun is assigned a gender according to each language's grammar rul…
In this work, we propose an analysis of the presence of gender bias associated with professions in Portuguese word embeddings. The objective of this work is to study gender implications related to stereotyped professions for women and men in the context of the Portuguese language.
IBM Research AI-India（IBM印度研究院人工智能部门）；IIIT-Delhi（德里英德拉普拉斯塔信息技术学院）；DTU-Delhi（德里理工大学）
The presence of gender stereotypes in many aspects of society is a well-known phenomenon. In this paper, we focus on studying and quantifying such stereotypes and bias in the Man Bookers Prize winning fiction. We consider 275 books shortlisted for Man Bookers Prize between 1969 and 2017. The gender bias is analyzed by semantic modeling of book descriptions on Goodreads. This reveals the pervasiveness of gender bias and stereotype in the books on different features like occupation, introductions and actions associated to the characters in the book.
The size of pretrained models is increasing, and so is their performance on a variety of NLP tasks. However, as their memorization capacity grows, they might pick up more social biases. In this work, we examine the connection between model size and its gender bias (specifically, occupational gender bias). We measure bias in three masked language model families (RoBERTa, DeBERTa, and T5) in two setups: directly using prompt based method, and using a downstream task (Winogender). We find on the one hand that larger models receive higher bias scores on the former task, but when evaluated on the l…
University of Groningen（格罗宁根大学）；University of Malta（马耳他大学）
Contextualized word embeddings have been replacing standard embeddings as the representational knowledge source of choice in NLP systems. Since a variety of biases have previously been found in standard word embeddings, it is crucial to assess biases encoded in their replacements as well. Focusing on BERT (Devlin et al., 2018), we measure gender bias by studying associations between gender-denoting target words and names of professions in English and German, comparing the findings with real-world workforce statistics. We mitigate bias by fine-tuning BERT on the GAP corpus (Webster et al., 2018…
Imagine a princess asleep in a castle, waiting for her prince to slay the dragon and rescue her. Tales like the famous Sleeping Beauty clearly divide up gender roles. But what about more modern stories, borne of a generation increasingly aware of social constructs like sexism and racism? Do these stories tend to reinforce gender stereotypes, or counter them? In this paper, we present a technique that combines natural language processing with a crowdsourced lexicon of stereotypes to capture gender biases in fiction. We apply this technique across 1.8 billion words of fiction from the Wattpad on…
The media attention to the personal sphere of famous and important individuals has become a key element of the gender narrative. Here we combine lexical, syntactic and sentiment analysis to investigate the role of gender in the personalization of a wide range of political office holders in Italy during the period 2017-2020. On the basis of a score for words that is introduced to account for gender unbalance in both representative and news coverage, we show that the political personalization in Italy is more detrimental for women than men, with the persistence of entrenched stereotypes includin…
This paper proposes two intuitive metrics, skew and stereotype, that quantify and analyse the gender bias present in contextual language models when tackling the WinoBias pronoun resolution task. We find evidence that gender stereotype correlates approximately negatively with gender skew in out-of-the-box models, suggesting that there is a trade-off between these two forms of bias. We investigate two methods to mitigate bias. The first approach is an online method which is effective at removing skew at the expense of stereotype. The second, inspired by previous work on ELMo, involves the fine-…
Boston University（波士顿大学）；Microsoft Research New England（微软新英格兰研究院）
Machine learning algorithms are optimized to model statistical properties of the training data. If the input data reflects stereotypes and biases of the broader society, then the output of the learning algorithm also captures these stereotypes. In this paper, we initiate the study of gender stereotypes in {\em word embedding}, a popular framework to represent text data. As their use becomes increasingly common, applications can inadvertently amplify unwanted stereotypes. We show across multiple datasets that the embeddings contain significant gender stereotypes, especially with regard to profe…
Studies have shown that some Natural Language Processing (NLP) systems encode and replicate harmful biases with potential adverse ethical effects in our society. In this article, we propose an approach for identifying gender and racial stereotypes in word embeddings trained on judicial opinions from U.S. case law. Embeddings containing stereotype information may cause harm when used by downstream systems for classification, information extraction, question answering, or other machine learning systems used to build legal research tools. We first explain how previously proposed methods for ident…
Wharton School, University of Pennsylvania（宾夕法尼亚大学沃顿商学院）
Cultural items like songs have an important impact in creating and reinforcing stereotypes, biases, and discrimination. But the actual nature of such items is often less transparent. Take songs, for example. Are lyrics biased against women? And how have any such biases changed over time? Natural language processing of a quarter of a million songs over 50 years quantifies misogyny. Women are less likely to be associated with desirable traits (i.e., competence), and while this bias has decreased, it persists. Ancillary analyses further suggest that song lyrics may help drive shifts in societal s…
As generic machine translation (MT) quality has improved, the need for targeted benchmarks that explore fine-grained aspects of quality has increased. In particular, gender accuracy in translation can have implications in terms of output fluency, translation accuracy, and ethics. In this paper, we introduce MT-GenEval, a benchmark for evaluating gender accuracy in translation from English into eight widely-spoken languages. MT-GenEval complements existing benchmarks by providing realistic, gender-balanced, counterfactual data in eight language pairs where the gender of individuals is unambiguo…
University of Mannheim（曼海姆大学）；University of Amsterdam（阿姆斯特丹大学）；University of Mannheim, GESIS, CSH Vienna（曼海姆大学、GESIS、维也纳CSH）；University of Hamburg（汉堡大学）
Assumed Identities: Quantifying Gender Bias in Machine Translation of Gender-Ambiguous Occupational Terms
School of Electrical and Computer Engineering, National Technical University of Athens（电气与计算机工程学院，国家技术大学雅典）；Instituto de Telecomunicações（电信研究所）；Department of Social Policy, Panteion University of Social and Political Sciences（社会政策系，帕内翁社会与政治科学大学）
Word embeddings use vectors to represent words such that the geometry between vectors captures semantic relationship between the words. In this paper, we develop a framework to demonstrate how the temporal dynamics of the embedding can be leveraged to quantify changes in stereotypes and attitudes toward women and ethnic minorities in the 20th and 21st centuries in the United States. We integrate word embeddings trained on 100 years of text data with the U.S. Census to show that changes in the embedding track closely with demographic and occupation shifts over time. The embedding captures globa…
National Research Council of Italy（意大利国家研究委员会）；Institute of Intelligent Systems for Automation – Genoa branch（智能系统自动化研究所热那亚分部）；Institute of Electronics, Computer and Telecommunication Engineering – Genoa branch（电子、计算机与通信工程研究所热那亚分部）
The purpose of this study is to find evidence for supporting the hypothesis that language is the mirror of our thinking, our prejudices and cultural stereotypes. In this analysis, a questionnaire was administered to 537 people. The answers have been analysed to see if gender stereotypes were present such as the attribution of psychological and behavioural characteristics. In particular, the aim was to identify, if any, what are the stereotyped images, which emerge in defining the roles of men and women in modern society. Moreover, the results given can be a good starting point to understand if…
International Institute of Information Technology, Bangalore（班加罗尔国际信息技术学院）；University of Gothenburg（哥德堡大学）；Chalmers University of Technology（查尔姆斯理工大学）
We propose a methodology and design two benchmark sets for measuring to what extent language-and-vision language models use the visual signal in the presence or absence of stereotypes. The first benchmark is designed to test for stereotypical colors of common objects, while the second benchmark considers gender stereotypes. The key idea is to compare predictions when the image conforms to the stereotype to predictions when it does not. Our results show that there is significant variation among multimodal models: the recent Transformer-based FLAVA seems to be more sensitive to the choice of ima…
Walmart Labs（沃尔玛实验室）；SAP Labs（思爱普实验室）；McGill University（麦吉尔大学）；Facebook AI Research（脸书人工智能研究）
Gender-bias stereotypes have recently raised significant ethical concerns in natural language processing. However, progress in detection and evaluation of gender bias in natural language understanding through inference is limited and requires further investigation. In this work, we propose an evaluation methodology to measure these biases by constructing a challenge task that involves pairing gender-neutral premises against a gender-specific hypothesis. We use our challenge task to investigate state-of-the-art NLI models on the presence of gender stereotypes using occupations. Our findings sug…
Seeds of Stereotypes: A Large-Scale Textual Analysis of Race and Gender Associations with Diseases in Online Sources
Aarhus University（奥胡斯大学）；Massachusetts Institute of Technology（麻省理工学院）；University of Alberta（阿尔伯塔大学）；University of Manitoba（曼尼托巴大学）；Mbarara University of Science and Technology（姆巴拉拉科技大学）；Fundacion Arturo Lopez Perez（阿图罗·洛佩斯·佩雷斯基金会）；Mass General Brigham（麻省总医院布里格姆）；Harvard Medical School（哈佛医学院）；Dana-Farber Cancer Institute（丹娜-法伯癌症研究所）；Emory University School of Medicine（埃默里大学医学院）；Beth Israel Deaconess Medical Center（贝斯以色列女执事医疗中心）；Harvard T.H. Chan School of Public Health（哈佛大学陈曾熙公共卫生学院）
本文通过分析 Arxiv、Wikipedia 和 Common Crawl 中18种疾病与种族、性别术语的共现，并对比患病率与 GPT-4 输出，揭示在线文本及 LLM 预训练数据中的人口统计学偏见。
GOSt-MT: A Knowledge Graph for Occupation-related Gender Biases in Machine Translation
National Technical University of Athens（雅典国立技术大学）；Panteion University of Social and Political Sciences（潘泰翁社会政治大学）；National and Kapodistrian University of Athens（雅典国立卡波迪斯特里安大学）
University of Oxford（牛津大学）；TU Wien（维也纳工业大学）；University College London（伦敦大学学院）
Bias-measuring datasets play a critical role in detecting biased behavior of language models and in evaluating progress of bias mitigation methods. In this work, we focus on evaluating gender bias through coreference resolution, where previous datasets are either hand-crafted or fail to reliably measure an explicitly defined bias. To overcome these shortcomings, we propose a novel method to collect diverse, natural, and minimally distant text pairs via counterfactual generation, and construct Counter-GAP, an annotated dataset consisting of 4008 instances grouped into 1002 quadruples. We furthe…
