Challenge: Existing work has established that a person’s demographics and speech style affect how well speech processing models perform for them.
Approach: They propose a method to detect bias in pre-trained models by using word embedding association tests in natural language processing to quantify bias in models' representations of different concepts.
Outcome: The proposed method detects bias in pre-trained models and can have real-world effects.

Similar Papers

StereoSet: Measuring stereotypical bias in pretrained language models (2021.acl-long)

Copied to clipboard

Challenge: Existing literature on stereotypical biases in language models is limited . current evaluations focus on measuring bias without considering language modeling ability .
Approach: They propose to measure stereotypical biases in four domains: gender, profession, race, and religion . they compare stereotypical and language modeling ability of popular models like BERT, GPT-2, RoBERTa and XLnet .
Outcome: The proposed model shows strong stereotypical biases in gender, profession, race, and religion domains.
Mind Your Bias: A Critical Review of Bias Detection Methods for Contextual Language Models (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for detection of biases in contextual language models are inconsistent and inconclusive.
Approach: They propose to use word embedding association test to detect biases in contextual language models to compare them with other methods.
Outcome: The proposed methods are inconsistent and inconclusive for language models with word embeddings.
From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models (2023.acl-long)

Copied to clipboard

Challenge: Hundreds of studies have highlighted ethical issues in NLP models .
Approach: They propose to measure media biases in LMs trained on diverse data sources . they focus on hate speech and misinformation detection .
Outcome: The proposed methods quantify the fairness of downstream NLP models trained on politically biased LMs.
On Measuring Social Biases in Sentence Encoders (N19-1)

Copied to clipboard

Challenge: Word embeddings such as word2vec and GloVe exhibit human-like implicit biases based on gender, race, and other social constructs.
Approach: They propose a simple generaliza test to measure bias in word embeddings by comparing two sets of target-concept words to two sets .
Outcome: The proposed test shows that word2vec and word2Ve exhibit human-like implicit biases based on gender, race, and other social constructs.
T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation (2023.findings-acl)

Copied to clipboard

Challenge: Recent advances in text-to-image generative models have produced high quality images with a breakthrough of inference speed.
Approach: They propose a text-to-image association test framework that quantifies implicit stereotypes between concepts and valence and those in images.
Outcome: The proposed framework quantifies implicit stereotypes between concepts and valence and those in images.
Upstream Mitigation Is Not All You Need: Testing the Bias Transfer Hypothesis in Pre-Trained Language Models (2022.acl-long)

Copied to clipboard

Challenge: Large language models and other massively pre-trained "foundation" models can easily adapt to a wide variety of downstream tasks in a process called finetuning.
Approach: They propose to use the bias transfer hypothesis to reduce social biases internalized by large language models during pre-training into harmful task-specific behavior after fine-tuning.
Outcome: The bias transfer hypothesis is the theory that social biases internalized by large language models during pre-training transfer into harmful task-specific behavior after fine-tuning.
The (Undesired) Attenuation of Human Biases by Multilinguality (2022.emnlp-main)

Copied to clipboard

Challenge: odor pleasantness perception is universal, but cultural biases are not always present in embedding models . et al., 2018: a new study shows that cultural bias is not always the case in embedded models based on human texts .
Approach: They propose multilingual cultural aware tests to quantify biases in embedding models . they find that biased models are more likely to be multilingual than monolingual ones .
Outcome: The results show that human preferences are not always universal . they also show that multilinguality reverses biases, despite differences in training corpus .
The Risk of Racial Bias in Hate Speech Detection (P19-1)

Copied to clipboard

Challenge: Annotators’ insensitivity to differences in dialect can lead to racial bias in automatic hate speech detection models, potentially amplifying harm against minority populations.
Approach: They propose *dialect* and *race priming* as ways to reduce the racial bias in hate speech detection models by detecting differences in dialects in annotated tweets.
Outcome: The proposed models acquire and propagate these biases, such that AAE tweets and tweets by self-identified African Americans are up to two times more likely to be labelled as offensive compared to others.
A Study of Implicit Bias in Pretrained Language Models against People with Disabilities (2022.coling-1)

Copied to clipboard

Challenge: Pretrained language models exhibit sociodemographic biases, such as against gender and race, raising concerns of downstream biase in language technologies.
Approach: They propose to use word embedding-based and transformer-based PLMs to test for the presence of biases against people with disabilities (PWDs)
Outcome: The proposed models favor ableist language, despite their sociodemographic biases against race and gender.
Marked Attribute Bias in Natural Language Inference (2021.findings-acl)

Copied to clipboard

Challenge: Existing tests for gender-biased word embeddings do not address marked attribute bias . authors propose a new type of intrinsic bias measure for static word embeds .
Approach: They propose a method to detect gender-biased word embeddings in a downstream NLP application . they propose 'debiasing' method to measure the marked attribute bias in embeddable word embeds .
Outcome: The proposed method achieves best results on the marked attribute bias test set.

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