Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks
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, \textbf{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.
Paper Type: Short
Research Area: Resources and Evaluation
Research Area Keywords: automatic evaluation of datasets, robustness, clinical NLP
Contribution Types: NLP engineering experiment, Data resources
Languages Studied: English
Submission Number: 1394
Loading