Affiliation:
1. University of Groningen, Center for Language and Cognition.
2. IT University of Copenhagen, Computer Science Department.
Abstract
Analogies such as man is to king as woman is to X are often used to illustrate the amazing power of word embeddings. Concurrently, they have also been used to expose how strongly human biases are encoded in vector spaces trained on natural language, with examples like man is to computer programmer as woman is to homemaker. Recent work has shown that analogies are in fact not an accurate diagnostic for bias, but this does not mean that they are not used anymore, or that their legacy is fading. Instead of focusing on the intrinsic problems of the analogy task as a bias detection tool, we discuss a series of issues involving implementation as well as subjective choices that might have yielded a distorted picture of bias in word embeddings. We stand by the truth that human biases are present in word embeddings, and, of course, the need to address them. But analogies are not an accurate tool to do so, and the way they have been most often used has exacerbated some possibly non-existing biases and perhaps hidden others. Because they are still widely popular, and some of them have become classics within and outside the NLP community, we deem it important to provide a series of clarifications that should put well-known, and potentially new analogies, into the right perspective.
Subject
Artificial Intelligence,Computer Science Applications,Linguistics and Language,Language and Linguistics
Cited by
27 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Asymmetric Similarity Scores of Color Names in Twitter-Based Language Models;2023 IEEE 13th International Workshop on Computational Intelligence and Applications (IWCIA);2023-11-11
2. Mitigate Gender Bias Using Negative Multi-task Learning;Neural Processing Letters;2023-07-28
3. From Word Types to Tokens and Back: A Survey of Approaches to Word Meaning Representation and Interpretation;Computational Linguistics;2023-03-14
4. The Political Biases of ChatGPT;Social Sciences;2023-03-02
5. Induction of Bilingual Dictionaries;Building and Using Comparable Corpora for Multilingual Natural Language Processing;2023