• Publications
  • Influence
Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Them
Word embeddings are widely used in NLP for a vast range of tasks. It was shown that word embeddings derived from text corpora reflect gender biases in society, causing serious concern. Several recent
Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection
TLDR
This work presents Iterative Null-space Projection (INLP), a novel method for removing information from neural representations based on repeated training of linear classifiers that predict a certain property the authors aim to remove, followed by projection of the representations on their null-space.
Simple, Interpretable and Stable Method for Detecting Words with Usage Change across Corpora
TLDR
This work proposes an alternative approach that does not use vector space alignment, and instead considers the neighbors of each word, and demonstrates its effectiveness in 9 different setups, considering different corpus splitting criteria.
It’s All in the Name: Mitigating Gender Bias with Name-Based Counterfactual Data Substitution
TLDR
CDA/S with the Names Intervention is the only approach which is able to mitigate indirect gender bias: following debiasing, previously biased words are significantly less clustered according to gender, thus improving on the state-of-the-art for bias mitigation.
Semi Supervised Preposition-Sense Disambiguation using Multilingual Data
TLDR
It is shown that signals from unannotated multilingual data can be used to improve supervised preposition-sense disambiguation, and pre-trains an LSTM encoder for predicting the translation of a preposition, and incorporates the pre-trained encoder as a component in a supervised classification system.
Automatically Identifying Gender Issues in Machine Translation using Perturbations
TLDR
A novel technique is developed to mine examples from real world data to explore challenges for deployed systems and expose where model representations are gendered, and the unintended consequences these gendered representations can have in downstream application.
Language Modeling for Code-Switching: Evaluation, Integration of Monolingual Data, and Discriminative Training
TLDR
An ASR-motivated evaluation setup is proposed which is decoupled from an ASR system and the choice of vocabulary, and this setup lends itself to a discriminative training approach, which is demonstrated to work better than generative language modeling.
It’s not Greek to mBERT: Inducing Word-Level Translations from Multilingual BERT
TLDR
The hypothesis that multilingual BERT learns representations which contain both a language-encoding component and an abstract, cross-lingual component is tested, and an empirical language-identity subspace within mBERT representations is identified.
How does Grammatical Gender Affect Noun Representations in Gender-Marking Languages?
TLDR
It is demonstrated that a careful application of methods that neutralize grammatical gender signals from the words’ context when training word embeddings is effective in removing it.
Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Them
TLDR
It is concluded that existing bias removal techniques are insufficient, and should not be trusted for providing gender-neutral modeling, for two debiasing methods.
...
1
2
...