Unmasking Contextual Stereotypes: Measuring and Mitigating BERT's Gender Bias

2020-10-27 · Marion Bartl, Malvina Nissim, Albert Gatt

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), after applying Counterfactual Data Substitution (CDS) (Maudslay et al., 2019). We show that our method of measuring bias is appropriate for languages such as English, but not for languages with a rich morphology and gender-marking, such as German. Our results highlight the importance of investigating bias and mitigation techniques cross-linguistically, especially in view of the current emphasis on large-scale, multilingual language models.

📄 PDF Abstract BibTeX arXiv:2010.14534

Code (1)

marionbartl/gender-bias-BERT 공식 구현 pytorch

Tasks

counterfactualWord Embeddings

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Multi-Head Attention 설명 없음

Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

WordPiece 설명 없음

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Adam 설명 없음

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Unmasking Contextual Stereotypes: Measuring and Mitigating BERT’s Gender Bias

2020-12-01 · GeBNLP (COLING) 2020 12 · Marion Bartl, Malvina Nissim, Albert Gatt

counterfactual Word Embeddings

On Measuring and Mitigating Biased Inferences of Word Embeddings

2019-08-25 · Sunipa Dev, Tao Li, Jeff Phillips, Vivek Srikumar

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…

Natural Language Inference Word Embeddings

Measuring Bias in Contextualized Word Representations

2019-06-18 · WS 2019 8 · Keita Kurita, Nidhi Vyas, Ayush Pareek, Alan W. black 외

Contextual word embeddings such as BERT have achieved state of the art performance in numerous NLP tasks. Since they are optimized to capture the statistical properties of training data, they tend to pick up on and ampli…

Word Embeddings

Dutch CrowS-Pairs: Adapting a Challenge Dataset for Measuring Social Biases in Language Models for Dutch

2025-07-22 · Elza Strazda, Gerasimos Spanakis arxiv

Warning: This paper contains explicit statements of offensive stereotypes which might be upsetting. Language models are prone to exhibiting biases, further amplifying unfair and harmful stereotypes. Given the fast-growin…

Controlling for Stereotypes in Multimodal Language Model Evaluation

2023-02-03 · Manuj Malik, Richard Johansson

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 …

Language Model Evaluation Language Modeling Language Modelling model