Projective Methods for Mitigating Gender Bias in Pre-trained Language Models

• URL: http://arxiv.org/abs/2403.18803v1
• Date: Wed, 27 Mar 2024 17:49:31 GMT
• Title: Projective Methods for Mitigating Gender Bias in Pre-trained Language Models
• Authors: Hillary Dawkins, Isar Nejadgholi, Daniel Gillis, Judi McCuaig,
• Abstract summary: Projective methods are fast to implement, use a small number of saved parameters, and make no updates to the existing model parameters. We find that projective methods can be effective at both intrinsic bias and downstream bias mitigation, but that the two outcomes are not necessarily correlated.
• Score: 10.418595661963062
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mitigation of gender bias in NLP has a long history tied to debiasing static word embeddings. More recently, attention has shifted to debiasing pre-trained language models. We study to what extent the simplest projective debiasing methods, developed for word embeddings, can help when applied to BERT's internal representations. Projective methods are fast to implement, use a small number of saved parameters, and make no updates to the existing model parameters. We evaluate the efficacy of the methods in reducing both intrinsic bias, as measured by BERT's next sentence prediction task, and in mitigating observed bias in a downstream setting when fine-tuned. To this end, we also provide a critical analysis of a popular gender-bias assessment test for quantifying intrinsic bias, resulting in an enhanced test set and new bias measures. We find that projective methods can be effective at both intrinsic bias and downstream bias mitigation, but that the two outcomes are not necessarily correlated. This finding serves as a warning that intrinsic bias test sets, based either on language modeling tasks or next sentence prediction, should not be the only benchmark in developing a debiased language model.

Related papers

• Unlabeled Debiasing in Downstream Tasks via Class-wise Low Variance Regularization [13.773597081543185]
  We introduce a novel debiasing regularization technique based on the class-wise variance of embeddings. Our method does not require attribute labels and targets any attribute, thus addressing the shortcomings of existing debiasing methods.
  arXiv  Detail & Related papers  (2024-09-29T03:56:50Z)
• Editable Fairness: Fine-Grained Bias Mitigation in Language Models [52.66450426729818]
  We propose a novel debiasing approach, Fairness Stamp (FAST), which enables fine-grained calibration of individual social biases. FAST surpasses state-of-the-art baselines with superior debiasing performance. This highlights the potential of fine-grained debiasing strategies to achieve fairness in large language models.
  arXiv  Detail & Related papers  (2024-08-07T17:14:58Z)
• Mitigating Biases for Instruction-following Language Models via Bias Neurons Elimination [54.865941973768905]
  We propose a novel and practical bias mitigation method, CRISPR, to eliminate bias neurons of language models in instruction-following settings. CRISPR automatically determines biased outputs and categorizes neurons that affect the biased outputs as bias neurons using an explainability method. Experimental results demonstrate the effectiveness of our method in mitigating biases under zero-shot instruction-following settings without losing the model's task performance and existing knowledge.
  arXiv  Detail & Related papers  (2023-11-16T07:16:55Z)
• Balancing Unobserved Confounding with a Few Unbiased Ratings in Debiased Recommendations [4.960902915238239]
  We propose a theoretically guaranteed model-agnostic balancing approach that can be applied to any existing debiasing method. The proposed approach makes full use of unbiased data by alternatively correcting model parameters learned with biased data, and adaptively learning balance coefficients of biased samples for further debiasing.
  arXiv  Detail & Related papers  (2023-04-17T08:56:55Z)
• Debiasing Stance Detection Models with Counterfactual Reasoning and Adversarial Bias Learning [15.68462203989933]
  Stance detection models tend to rely on dataset bias in the text part as a shortcut. We propose an adversarial bias learning module to model the bias more accurately.
  arXiv  Detail & Related papers  (2022-12-20T16:20:56Z)
• Feature-Level Debiased Natural Language Understanding [86.8751772146264]
  Existing natural language understanding (NLU) models often rely on dataset biases to achieve high performance on specific datasets. We propose debiasing contrastive learning (DCT) to mitigate biased latent features and neglect the dynamic nature of bias. DCT outperforms state-of-the-art baselines on out-of-distribution datasets while maintaining in-distribution performance.
  arXiv  Detail & Related papers  (2022-12-11T06:16:14Z)
• MABEL: Attenuating Gender Bias using Textual Entailment Data [20.489427903240017]
  We propose MABEL, an intermediate pre-training approach for mitigating gender bias in contextualized representations. Key to our approach is the use of a contrastive learning objective on counterfactually augmented, gender-balanced entailment pairs. We show that MABEL outperforms previous task-agnostic debiasing approaches in terms of fairness.
  arXiv  Detail & Related papers  (2022-10-26T18:36:58Z)
• Gender Biases and Where to Find Them: Exploring Gender Bias in Pre-Trained Transformer-based Language Models Using Movement Pruning [32.62430731115707]
  We show a novel framework for inspecting bias in transformer-based language models via movement pruning. We implement our framework by pruning the model while fine-tuning it on the debiasing objective. We re-discover a bias-performance trade-off: the better the model performs, the more bias it contains.
  arXiv  Detail & Related papers  (2022-07-06T06:20:35Z)
• Balancing out Bias: Achieving Fairness Through Training Reweighting [58.201275105195485]
  Bias in natural language processing arises from models learning characteristics of the author such as gender and race. Existing methods for mitigating and measuring bias do not directly account for correlations between author demographics and linguistic variables. This paper introduces a very simple but highly effective method for countering bias using instance reweighting.
  arXiv  Detail & Related papers  (2021-09-16T23:40:28Z)
• Double Perturbation: On the Robustness of Robustness and Counterfactual Bias Evaluation [109.06060143938052]
  We propose a "double perturbation" framework to uncover model weaknesses beyond the test dataset. We apply this framework to study two perturbation-based approaches that are used to analyze models' robustness and counterfactual bias in English.
  arXiv  Detail & Related papers  (2021-04-12T06:57:36Z)
• Towards Robustifying NLI Models Against Lexical Dataset Biases [94.79704960296108]
  This paper explores both data-level and model-level debiasing methods to robustify models against lexical dataset biases. First, we debias the dataset through data augmentation and enhancement, but show that the model bias cannot be fully removed via this method. The second approach employs a bag-of-words sub-model to capture the features that are likely to exploit the bias and prevents the original model from learning these biased features.
  arXiv  Detail & Related papers  (2020-05-10T17:56:10Z)

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.