Fewer Errors, but More Stereotypes? The Effect of Model Size on Gender Bias

• URL: http://arxiv.org/abs/2206.09860v1
• Date: Mon, 20 Jun 2022 15:52:40 GMT
• Title: Fewer Errors, but More Stereotypes? The Effect of Model Size on Gender Bias
• Authors: Yarden Tal, Inbal Magar, Roy Schwartz
• Abstract summary: We examine the connection between model size and its gender bias. We find on the one hand that larger models receive higher bias scores on the former task, but when evaluated on the latter, they make fewer gender errors.
• Score: 5.077090615019091
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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 latter, they make fewer gender errors. To examine these potentially conflicting results, we carefully investigate the behavior of the different models on Winogender. We find that while larger models outperform smaller ones, the probability that their mistakes are caused by gender bias is higher. Moreover, we find that the proportion of stereotypical errors compared to anti-stereotypical ones grows with the model size. Our findings highlight the potential risks that can arise from increasing model size.

Related papers

• Investigating Gender Bias in LLM-Generated Stories via Psychological Stereotypes [8.091664636677637]
  We investigate gender bias in Large Language Models (LLMs) using gender stereotypes studied in psychology.<n>We introduce a novel dataset called StereoBias-Stories containing short stories either unconditioned or conditioned on (one, two, or six) random attributes from 25 psychological stereotypes.<n>We analyze how the gender contribution in the overall story changes in response to these attributes and present three key findings.
  arXiv  Detail & Related papers  (2025-08-05T10:10:26Z)
• Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Models [49.41113560646115]
  We investigate various proxy measures of bias in large language models (LLMs)<n>We find that evaluating models with pre-prompted personae on a multi-subject benchmark (MMLU) leads to negligible and mostly random differences in scores.<n>With the recent trend for LLM assistant memory and personalization, these problems open up from a different angle.
  arXiv  Detail & Related papers  (2025-06-12T08:47:40Z)
• Adultification Bias in LLMs and Text-to-Image Models [55.02903075972816]
  We study bias along axes of race and gender in young girls.<n>We focus on "adultification bias," a phenomenon in which Black girls are presumed to be more defiant, sexually intimate, and culpable than their White peers.
  arXiv  Detail & Related papers  (2025-06-08T21:02:33Z)
• Do Existing Testing Tools Really Uncover Gender Bias in Text-to-Image Models? [11.101062595569854]
  Previous studies have shown that Text-to-Image (T2I) models can perpetuate or even amplify gender stereotypes when provided with neutral text prompts. No existing work comprehensively compares the various detectors and understands how the gender bias detected by them deviates from the actual situation. This study addresses this gap by validating previous gender bias detectors using a manually labeled dataset and comparing how the bias identified by various detectors deviates from the actual bias in T2I models.
  arXiv  Detail & Related papers  (2025-01-27T04:47:19Z)
• Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs) [82.57490175399693]
  We study gender bias in 22 popular image-to-text vision-language assistants (VLAs) Our results show that VLAs replicate human biases likely present in the data, such as real-world occupational imbalances. To eliminate the gender bias in these models, we find that fine-tuning-based debiasing methods achieve the best trade-off between debiasing and retaining performance.
  arXiv  Detail & Related papers  (2024-10-25T05:59:44Z)
• The Male CEO and the Female Assistant: Evaluation and Mitigation of Gender Biases in Text-To-Image Generation of Dual Subjects [58.27353205269664]
  We propose the Paired Stereotype Test (PST) framework, which queries T2I models to depict two individuals assigned with male-stereotyped and female-stereotyped social identities. PST queries T2I models to depict two individuals assigned with male-stereotyped and female-stereotyped social identities. Using PST, we evaluate two aspects of gender biases -- the well-known bias in gendered occupation and a novel aspect: bias in organizational power.
  arXiv  Detail & Related papers  (2024-02-16T21:32:27Z)
• Will the Prince Get True Love's Kiss? On the Model Sensitivity to Gender Perturbation over Fairytale Texts [87.62403265382734]
  Recent studies show that traditional fairytales are rife with harmful gender biases. This work aims to assess learned biases of language models by evaluating their robustness against gender perturbations.
  arXiv  Detail & Related papers  (2023-10-16T22:25:09Z)
• Exploring Gender Bias in Retrieval Models [2.594412743115663]
  Mitigating gender bias in information retrieval is important to avoid propagating stereotypes. We employ a dataset consisting of two components: (1) relevance of a document to a query and (2) "gender" of a document. We show that pre-trained models for IR do not perform well in zero-shot retrieval tasks when full fine-tuning of a large pre-trained BERT encoder is performed. We also illustrate that pre-trained models have gender biases that result in retrieved articles tending to be more often male than female.
  arXiv  Detail & Related papers  (2022-08-02T21:12:05Z)
• Predicting on the Edge: Identifying Where a Larger Model Does Better [61.793778186198864]
  We show that large models have the largest improvement on examples where the small model is most uncertain. We show that a switcher model which defers examples to a larger model when a small model is uncertain can achieve striking improvements in performance and resource usage.
  arXiv  Detail & Related papers  (2022-02-15T18:53:14Z)
• Are Commercial Face Detection Models as Biased as Academic Models? [64.71318433419636]
  We compare academic and commercial face detection systems, specifically examining robustness to noise. We find that state-of-the-art academic face detection models exhibit demographic disparities in their noise robustness. We conclude that commercial models are always as biased or more biased than an academic model.
  arXiv  Detail & Related papers  (2022-01-25T02:21:42Z)
• Collecting a Large-Scale Gender Bias Dataset for Coreference Resolution and Machine Translation [10.542861450223128]
  We find grammatical patterns indicating stereotypical and non-stereotypical gender-role assignments in corpora from three domains. We manually verify the quality of our corpus and use it to evaluate gender bias in various coreference resolution and machine translation models.
  arXiv  Detail & Related papers  (2021-09-08T18:14:11Z)
• Exploring Biases and Prejudice of Facial Synthesis via Semantic Latent Space [1.858151490268935]
  This work targets biased generative models' behaviors, identifying the cause of the biases and eliminating them. We can (as expected) conclude that biased data causes biased predictions of face frontalization models. We found that the seemingly obvious choice of 50:50 proportions was not the best for this dataset to reduce biased behavior on female faces.
  arXiv  Detail & Related papers  (2021-08-23T16:09:18Z)
• Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Models [5.378664454650768]
  This paper proposes two intuitive metrics, skew and stereotype, that quantify and analyse the gender bias present in contextual language models. 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.
  arXiv  Detail & Related papers  (2021-01-24T10:57:59Z)
• UnQovering Stereotyping Biases via Underspecified Questions [68.81749777034409]
  We present UNQOVER, a framework to probe and quantify biases through underspecified questions. We show that a naive use of model scores can lead to incorrect bias estimates due to two forms of reasoning errors. We use this metric to analyze four important classes of stereotypes: gender, nationality, ethnicity, and religion.
  arXiv  Detail & Related papers  (2020-10-06T01:49:52Z)
• Mitigating Gender Bias in Captioning Systems [56.25457065032423]
  Most captioning models learn gender bias, leading to high gender prediction errors, especially for women. We propose a new Guided Attention Image Captioning model (GAIC) which provides self-guidance on visual attention to encourage the model to capture correct gender visual evidence.
  arXiv  Detail & Related papers  (2020-06-15T12:16:19Z)
• Do Neural Ranking Models Intensify Gender Bias? [13.37092521347171]
  We first provide a bias measurement framework which includes two metrics to quantify the degree of the unbalanced presence of gender-related concepts in a given IR model's ranking list. Applying these queries to the MS MARCO Passage retrieval collection, we then measure the gender bias of a BM25 model and several recent neural ranking models. Results show that while all models are strongly biased toward male, the neural models, and in particular the ones based on contextualized embedding models, significantly intensify gender bias.
  arXiv  Detail & Related papers  (2020-05-01T13:31:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.