Evaluating Short-Term Temporal Fluctuations of Social Biases in Social Media Data and Masked Language Models

Yi Zhou, Danushka Bollegala, Jose Camacho-Collados


Abstract
Social biases such as gender or racial biases have been reported in language models (LMs), including Masked Language Models (MLMs). Given that MLMs are continuously trained with increasing amounts of additional data collected over time, an important yet unanswered question is how the social biases encoded with MLMs vary over time. In particular, the number of social media users continues to grow at an exponential rate, and it is a valid concern for the MLMs trained specifically on social media data whether their social biases (if any) would also amplify over time. To empirically analyse this problem, we use a series of MLMs pretrained on chronologically ordered temporal snapshots of corpora. Our analysis reveals that, although social biases are present in all MLMs, most types of social bias remain relatively stable over time (with a few exceptions). To further understand the mechanisms that influence social biases in MLMs, we analyse the temporal corpora used to train the MLMs. Our findings show that some demographic groups, such as male, obtain higher preference over the other, such as female on the training corpora constantly.
Anthology ID:
2024.emnlp-main.1098
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
19693–19708
Language:
URL:
https://aclanthology.org/2024.emnlp-main.1098
DOI:
Bibkey:
Cite (ACL):
Yi Zhou, Danushka Bollegala, and Jose Camacho-Collados. 2024. Evaluating Short-Term Temporal Fluctuations of Social Biases in Social Media Data and Masked Language Models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 19693–19708, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Evaluating Short-Term Temporal Fluctuations of Social Biases in Social Media Data and Masked Language Models (Zhou et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.1098.pdf
Software:
 2024.emnlp-main.1098.software.zip