Understanding Undesirable Word Embedding Associations
Word embeddings are often criticized for capturing undesirable word associations such as gender stereotypes. However, methods for measuring and removing such biases remain poorly understood. We show that for any embedding model that implicitly does matrix factorization, debiasing vectors post hoc using subspace projection (Bolukbasi et al., 2016) is, under certain conditions, equivalent to training on an unbiased corpus. We also prove that WEAT, the most common association test for word embeddings, systematically overestimates bias. Given that the subspace projection method is provably effective, we use it to derive a new measure of association called the $\textit{relational inner product association}$ (RIPA). Experiments with RIPA reveal that, on average, skipgram with negative sampling (SGNS) does not make most words any more gendered than they are in the training corpus. However, for gender-stereotyped words, SGNS actually amplifies the gender association in the corpus.
Code (0)
등록된 구현이 없습니다.
Tasks
Word EmbeddingsSimilar Papers 제목 키워드 기반
Grammatical gender associations outweigh topical gender bias in crosslinguistic word embeddings
Recent research has demonstrated that vector space models of semantics can reflect undesirable biases in human culture. Our investigation of crosslinguistic word embeddings reveals that topical gender bias interacts with…
Cultural Vocal Bursts Intensity PredictionLemmatizationMachine TranslationTranslation+1The Undesirable Dependence on Frequency of Gender Bias Metrics Based on Word Embeddings
Numerous works use word embedding-based metrics to quantify societal biases and stereotypes in texts. Recent studies have found that word embeddings can capture semantic similarity but may be affected by word frequency. …
Semantic SimilaritySemantic Textual SimilarityWord EmbeddingsUnmasking the Trade-off: Measuring Gender Bias Mitigation and Over-debiasing Effects in Pretrained Language Models
Pretrained language models (PLMs) have demonstrated success across many natural language processing tasks. However, evidence suggests that they encode gender bias present in the corpora they are trained on. Existing bias…
Model Choices Influence Attributive Word Associations: A Semi-supervised Analysis of Static Word Embeddings
Static word embeddings encode word associations, extensively utilized in downstream NLP tasks. Although prior studies have discussed the nature of such word associations in terms of biases and lexical regularities captur…
ClusteringWord EmbeddingsNurse is Closer to Woman than Surgeon? Mitigating Gender-Biased Proximities in Word Embeddings
Word embeddings are the standard model for semantic and syntactic representations of words. Unfortunately, these models have been shown to exhibit undesirable word associations resulting from gender, racial, and religiou…
coreference-resolutionCoreference ResolutionWord Embeddings