Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Function
Open accessThe DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
Gender bias exists in natural language datasets which neural language models tend to learn, resulting in biased text generation.In this research, we propose a debiasing approach based on the loss function modification.We introduce a new term to the loss function which attempts to equalize the probabilities of male and female words in the output.Using an array of bias evaluation metrics, we provide empirical evidence that our approach successfully mitigates gender bias in language models without increasing perplexity by much.In comparison to existing debiasing strategies, data augmentation, and word embedding debiasing, our method performs better in several aspects, especially in reducing gender bias in occupation words.Finally, we introduce a combination of data augmentation and our approach, and show that it outperforms existing strategies in all bias evaluation metrics.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V., & Kalai, A. (2016) · arXiv · doi:10.48550/arxiv.1607.06520 · Cited by 1,615 · cited 5×
“Bolukbasi et al. (2016) and Zhao et al. (2017) defined gender bias as a projection of gender-neutral words on a gender subspace and removed bias by minimizing this projection.”
BackgroundMethodInfluentialMen Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints
Zhao, J., Wang, T., Yatskar, M., Ordonez, V., & Chang, K.-W. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1323 · Cited by 829 · cited 6×
“Bolukbasi et al. (2016) and Zhao et al. (2017) defined gender bias as a projection of gender-neutral words on a gender subspace and removed bias by minimizing this projection.”
BackgroundMethodInfluentialGonen, H., & Goldberg, Y. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1061 · Cited by 454 · cited 3×
“But, this bias definition is shown to be incomplete by Gonen and Goldberg (2019).”
BackgroundInfluentialIdentifying and Reducing Gender Bias in Word-Level Language Models
Bordia, S., & Bowman, S. R. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop · doi:10.18653/v1/n19-3002 · Cited by 219
Cited by
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
“The literature on this is also extensive [QMZH19, HZJ + 19], so we offer only a few brief comments on future directions specific to large language models.”
BackgroundLanguage (Technology) is Power: A Critical Survey of “Bias” in NLP
Blodgett, S. L., Barocas, S., Daumé III, H., & Wallach, H. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.485 · Cited by 1,050
“…Maudslay et al. (2019); May et al. (2019); Mirzaev et al. (2019); Prates et al. (2019); Précenth (2019); Prost et al. (2019); Pujari et al. (2019); Qian et al. (2019); Sedoc and Ungar (2019); Stanovsky et al. (2019); Tan and Celis (2019); Zhao et al. (2019); Zhou et al. (2019); Chopra et al.…”
InfluentialBias and Fairness in Large Language Models: A Survey
Gallegos, I. O., Rossi, R. A., Barrow, J., Tanjim, M. M., Kim, S., Dernoncourt, F., Yu, T., Zhang, R., & Ahmed, N. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00524 · Cited by 1,245
A Survey of Controllable Text Generation Using Transformer-based Pre-trained Language Models
Zhang, H., Song, H., Li, S., Zhou, M., & Song, D. (2024) · ACM Computing Surveys · doi:10.1145/3617680 · Cited by 317
Towards Understanding and Mitigating Social Biases in Language Models
Liang, P. P., Wu, C., Morency, L.-P., & Salakhutdinov, R. (2021) · arXiv · doi:10.48550/arxiv.2106.13219 · Cited by 153
“Existing approaches towards mitigating biases in generation currently require retraining the models through adversarial trigger prompts (Sheng et al., 2020), data augmentation or collection (Dinan et al., 2020), and different objective functions (Qian et al., 2019; Huang et al., 2020).”
Background
Metadata
- Authors
- Qian, Y., Muaz, U., Zhang, B., & Hyun, J. W.
- Venue
- Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop
- Published
- 2019
- Pages
- 223-228
- OpenAlex
- W2963524349
- Cited by
- 63
- References
- 4
- FWCI
- 10.10
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.