Challenge: Recent research has exposed text encoders for replicating discriminatory social biases which may cause unintended and undesired model behaviors with respect to social groups.
Approach: They propose a method to reduce social stereotypes by redistributing attention scores of a text encoder so it forgets any preference to historically advantaged groups and attends to all social classes with the same intensity.
Outcome: The proposed method reduces stereotypes and inflicts no semantic damage on pre-trained encoders.

Similar Papers

Deciphering Stereotypes in Pre-Trained Language Models (2023.emnlp-main)

Copied to clipboard

Challenge: Current approaches for examining stereotypes in PLMs require intricate human knowledge about these stereotypes and entail careful manual curation of examples.
Approach: They propose a framework for examining stereotype-encoding behavior of PLMs using model probing and textual analyses.
Outcome: The proposed approach can debiase PLMs without compromising their language modeling capabilities or performance.
When Debiasing Backfires: Counterintuitive Side Effects of Preprocessing-Based Stereotype Mitigation (2026.findings-acl)

Copied to clipboard

Challenge: Preprocessing-based methods for stereotype mitigation are widely used in NLP . preprocessing methods cause unintended shifts in attention flow, authors say .
Approach: They propose to use preprocessing-based methods to reduce stereotypes for targeted groups . they find that stereotyping or counter-stereotyping can increase for other demographics .
Outcome: The proposed methods often induce unintended shifts across demographics, the authors show . they show that such side effects are not accompanied by large changes in attention flow .
Towards Debiasing Sentence Representations (2020.acl-main)

Copied to clipboard

Challenge: Recent work has shown word-level embeddings reflect and propagate social biases present in training corpora.
Approach: They propose a method to debias word embeddings to reduce biases at sentence level . they hope their work will inspire future research on characterizing and removing biase .
Outcome: The proposed method reduces biases and preserves performance on downstream tasks such as sentiment analysis and natural language understanding.
Debiasing Pre-trained Contextualised Embeddings (2021.eacl-main)

Copied to clipboard

Challenge: a study of contextualised word embeddings shows discriminative biases are encoded in contextualised embeddables.
Approach: They propose a fine-tuning method that can be applied at token- or sentence-levels to debias pre-trained contextualised embeddings.
Outcome: The proposed method can be applied at token- or sentence-levels to debias pre-trained models without requiring retrains.
Gender-preserving Debiasing for Pre-trained Word Embeddings (P19-1)

Copied to clipboard

Challenge: Existing methods for debiasing word embeddings have shown discriminative biases . word embeds learnt from social media have shown to encode racist, offensive and discriminative language usage.
Approach: They propose a method that preserves gender-related information while removing stereotypical gender biases from pre-trained word embeddings.
Outcome: The proposed method preserves gender-related information while removing stereotypical discriminative gender biases from pre-trained word embeddings.
Unlabeled Debiasing in Downstream Tasks via Class-wise Low Variance Regularization (2024.emnlp-main)

Copied to clipboard

Challenge: Existing methods for debiasing depend on attribute labels and target attributes.
Approach: They propose a method that uses class-wise variance of embeddings to reduce the effects of debiasing on a downstream task.
Outcome: The proposed method outperforms baselines that rely on attribute labels while maintaining performance on the target task.
From Prejudice to Parity: A New Approach to Debiasing Large Language Model Word Embeddings (2025.coling-main)

Copied to clipboard

Challenge: Existing work in this field has looked most commonly into gender bias, racial bias, and religious bias.
Approach: They propose an algorithm that uses a neural network to perform ‘soft debiasing’ and build on the seminal work of (CITATION) and (CitATION).
Outcome: The proposed algorithm outperforms current methods on gender, race, and religion metrics on a wide range of metrics.
Dynamically Disentangling Social Bias from Task-Oriented Representations with Adversarial Attack (2021.naacl-main)

Copied to clipboard

Challenge: Existing methods to learn representations from text often reflect social biases . previous methods rely on pre-specified direction or suffer from unstable training .
Approach: They propose an adversarial disentangled debiasing model to decouple social bias attributes from intermediate representations trained on the main task.
Outcome: The proposed model decouples social bias attributes from intermediate representations trained on the main task.
Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP (2021.tacl-1)

Copied to clipboard

Challenge: Pretrained language models pick up and reproduce undesirable biases when trained on large, unfiltered crawls from the Internet.
Approach: They propose a decoding algorithm that, given only a textual description of the undesired behavior, reduces the probability of a language model producing problematic text.
Outcome: The proposed approach reduces the probability of a language model producing problematic text by giving only a textual description of the undesired behavior.
Black is to Criminal as Caucasian is to Police: Detecting and Removing Multiclass Bias in Word Embeddings (N19-1)

Copied to clipboard

Challenge: Existing methods to debias word embeddings in binary settings such as gender and religion are limited to binary labels, whereas word2vec embedders can be used to propagate biases.
Approach: They propose a method to debias word embeddings in multiclass settings such as gender and religion, extending the work of Bolukbasi et al. (2016).
Outcome: The proposed method maintains the efficacy in standard NLP tasks while maintaining the utility of embeddings.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations