EUVIMEDEUVIMED entdecken
Crossref

Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Models

Shweta Parihar, Guangliang Liu, Natalie Parde, Lu Cheng

ACM SIGKDD Explorations Newsletter · 17.06.2026 · 28 (1) · 21-31

DOI: 10.1145/3820356.3820358 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM SIGKDD Explorations Newsletter, online erschienen am 17.06.2026. Verfasst von Shweta Parihar und 3 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A challenge in mitigating social bias in fine-tuned language models (LMs) is the potential reduction in language modeling capability, which can harm downstream performance. Counterfactual data augmentation (CDA), a widely used method for fine-tuning, highlights this issue by generating synthetic data that may align poorly with real-world distributions or creating overly simplistic counterfactuals that ignore the social context of altered sensitive attributes (e.g., gender) in the pretraining corpus. To address these limitations, we propose a simple yet effective context-augmented CDA method, Context-CDA, which uses large LMs to enhance the diversity and contextual relevance of the debiasing corpus. By minimizing discrepancies between the debiasing corpus and pretraining data through augmented context, this approach ensures better alignment, enhancing language modeling capability. We then employ uncertainty-based filtering to exclude generated counterfactuals considered low-quality by the target smaller LMs (i.e., LMs to be debiased), further improving the fine-tuning corpus quality. Experimental results on gender bias benchmarks demonstrate that Context-CDA effectively mitigates bias without sacrificing language modeling performance while offering insights into social biases by analyzing distribution shifts in next-token generation probabilities.

Zitieren

APA

Parihar, S., Liu, G., Parde, N., & Cheng, L. (2026). Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Models. ACM SIGKDD Explorations Newsletter, 28(1), 21-31. https://doi.org/10.1145/3820356.3820358

Vancouver

Parihar S, Liu G, Parde N, Cheng L. Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Models. ACM SIGKDD Explorations Newsletter. 2026;28(1):21-31. doi:10.1145/3820356.3820358

BibTeX
@article{Parihar2026,
  author = {Parihar, Shweta and Liu, Guangliang and Parde, Natalie and Cheng, Lu},
  title = {Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Models},
  journal = {ACM SIGKDD Explorations Newsletter},
  year = {2026},
  volume = {28},
  number = {1},
  pages = {21--31},
  doi = {10.1145/3820356.3820358}
}
Kostenlos starten →