"You are grounded!": Latent Name Artifacts in Pre-trained Language Models

Vered Shwartz, Rachel Rudinger, Oyvind Tafjord
2020 Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)   unpublished
Pre-trained language models (LMs) may perpetuate biases originating in their training corpus to downstream models. We focus on artifacts associated with the representation of given names (e.g., Donald), which, depending on the corpus, may be associated with specific entities, as indicated by next token prediction (e.g., Trump). While helpful in some contexts, grounding happens also in underspecified or inappropriate contexts. For example, endings generated for 'Donald is a' substantially differ
more » ... ubstantially differ from those of other names, and often have more-than-average negative sentiment. We demonstrate the potential effect on downstream tasks with reading comprehension probes where name perturbation changes the model answers. As a silver lining, our experiments suggest that additional pre-training on different corpora may mitigate this bias. Corinne A Moss-Racusin, John F Dovidio, Victoria L Brescoll, Mark J Graham, and Jo Handelsman. 2012. Science faculty's subtle gender biases favor male students. . 2019. Exploring the limits of transfer learning with a unified text-to-text transformer. ArXiv, abs/1910.10683. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. Squad: 100,000+ questions for machine comprehension of text. In EMNLP.
doi:10.18653/v1/2020.emnlp-main.556 fatcat:sxzrrxpsjnax5frgrfisoeytrm