Note: these datasets were uploaded on March 19th, 2024 by Hugging Face staff from the ancillary files attached to the original arXiv submission.

Supplementary Material for "A Toolbox for Surfacing Health Equity Harms and Biases in Large Language Models"

https://arxiv.org/abs/2403.12025

We include adversarial questions for each of the seven EquityMedQA datasets: OMAQ, EHAI, FBRT-Manual, FBRT-LLM, TRINDS, CC-Manual, and CC-LLM. For FBRT-LLM, we include both the full set and the subset we sampled for evaluation in the work. For CC-Manual and CC-LLM, we provide two related questions on each line in their respective files. Data generated as a part of the empirical study (Med-PaLM 2 model outputs and human ratings) are not included in EquityMedQA.

We also include other datasets evaluated in this work: MultiMedQA, Mixed MMQA-OMAQ, and Omiye et al. These datasets are derived from:

1. Singhal, K., Azizi, S., Tu, T. et al. Large language models encode clinical knowledge. Nature 620, 172–180 (2023). https://doi.org/10.1038/s41586-023-06291-2
2. Omiye, J.A., Lester, J.C., Spichak, S. et al. Large language models propagate race-based medicine. npj Digit. Med. 6, 195 (2023). https://doi.org/10.1038/s41746-023-00939-z

See the paper for details on all datasets.

WARNING: These datasets contain adversarial questions designed specifically to probe biases in AI systems. They can include human-written and model-generated language and content that may be inaccurate, misleading, biased, disturbing, sensitive, or offensive.

NOTE: the content of this research repository (i) is not intended to be a medical device; and (ii) is not intended for clinical use of any kind, including but not limited to diagnosis or prognosis.