Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks

Jack Gallifant, Shan Chen, Pedro Moreira, Nikolaj Munch, Mingye Gao, Jackson Pond, Leo Anthony Celi, Hugo J. W. L. Aerts, Thomas Hartvigsen, Danielle S. Bitterman

Abstract

Medical knowledge is context-dependent and requires consistent reasoning across various natural language expressions of semantically equivalent phrases. This is particularly crucial for drug names, where patients often use brand names like Advil or Tylenol instead of their generic equivalents. To study this, we create a new robustness dataset, RABBITS, to evaluate performance differences on medical benchmarks after swapping brand and generic drug names using physician expert annotations.We assess both open-source and API-based LLMs on MedQA and MedMCQA, revealing a consistent performance drop ranging from 1-10%. Furthermore, we identify a potential source of this fragility as the contamination of test data in widely used pre-training datasets.
BibTeX
@inproceedings{gallifant2024language,
  author = {Jack Gallifant and Shan Chen and Pedro Moreira and Nikolaj Munch and Mingye Gao and Jackson Pond and Leo Anthony Celi and Hugo J. W. L. Aerts and Thomas Hartvigsen and Danielle S. Bitterman},
  title = {{Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks}},
  booktitle = {Conference on Empirical Methods in Natural Language Processing},
  year = {2024},
  doi = {10.18653/v1/2024.findings-emnlp.726},
}
Figure