mlsanityData to 13 Oct 2026

Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks

arXiv 2406.12066 ↗First posted 17 Jun 2024cs.CLcited by 17

Not on the map. Why?

Abstract (arXiv)

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. All code is accessible at https://github.com/BittermanLab/RABBITS, and a HuggingFace leaderboard is available at https://huggingface.co/spaces/AIM-Harvard/rabbits-leaderboard.

Key facts. Cited by 17 papers in the map (data through 13 Oct 2026). Builds on 25 of its 52 references that are in the map. First posted on arXiv 17 Jun 2024.

Builds on

25 of 52 references.

References · 25

  1. 2020Language Models are Few-Shot Learnerscited by 32,320
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,549
  3. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,979
  4. 2020Measuring Massive Multitask Language Understandingcited by 7,876
  5. 2020Scaling Laws for Neural Language Modelscited by 6,494
  6. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,379
  7. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,908
  8. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,799
  9. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,510
  10. 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,338
  11. 2019PubMedQA: A Dataset for Biomedical Research Question Answeringcited by 1,281
  12. 2024Yi: Open Foundation Models by 01.AIcited by 715
13 more
  1. 2023Textbooks Are All You Needcited by 566
  2. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 564
  3. 2024Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Researchcited by 477
  4. 2023Don't Make Your LLM an Evaluation Benchmark Cheatercited by 194
  5. 2024A Careful Examination of Large Language Model Performance on Grade School Arithmeticcited by 191
  6. 2024Infini-gram: Scaling Unbounded n-gram Language Models to a Trillion Tokenscited by 149
  7. 2023Are Emergent Abilities in Large Language Models just In-Context Learning?cited by 109
  8. 2023Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Datasetcited by 90
  9. 2024Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gapcited by 79
  10. 2023Paloma: A Benchmark for Evaluating Language Model Fitcited by 53
  11. 2024Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQAcited by 19
  12. 2020MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretrainingcited by 18
  13. 2022Always Keep your Target in Mind: Studying Semantics and Improving Performance of Neural Lexical Substitutioncited by 12

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2024: 4; 2025: 8; 2026: 5.

17 papers.

Sort:

By area:

  1. 2025TxAgent: An AI Agent for Therapeutic Reasoning Across a Universe of Toolscited by 58
  2. 2025A Comprehensive Survey on the Trustworthiness of Large Language Models in Healthcarecited by 27
  3. 2024WorldMedQA-V: a multilingual, multimodal medical examination dataset for multimodal language models evaluationcited by 11
  4. 2024"Oh LLM, I'm Asking Thee, Please Give Me a Decision Tree": Zero-Shot Decision Tree Induction and Embedding with Large Language Modelscited by 6
  5. 2024EHRmonize: A Framework for Medical Concept Abstraction from Electronic Health Records using Large Language Modelscited by 4
  6. 2025Facts Fade Fast: Evaluating Memorization of Outdated Medical Knowledge in Large Language Modelscited by 4
  7. 2024Wait, but Tylenol is Acetaminophen... Investigating and Improving Language Models' Ability to Resist Requests for Misinformationcited by 2
  8. 2026How Does "English (US)" Become the Default? Triangulating Structural Bias Towards American English Across the LLM Pipelinecited by 2
  9. 2025Do "New Snow Tablets" Contain Snow? Large Language Models Over-Rely on Names to Identify Ingredients of Chinese Drugscited by 1
  10. 2025Testing for LLM response differences: the case of a composite null consisting of semantically irrelevant query perturbationscited by 1
  11. 2025How Well Do LLMs Understand Drug Mechanisms? A Knowledge + Reasoning Evaluation Datasetcited by 1
  12. 2025Towards medical AI misalignment: a preliminary studycited by 0
5 more
  1. 2025A Field Guide to Deploying AI Agents in Clinical Practicecited by 0
  2. 2026What's in a Name? Morphological Shortcuts by LLMs in Pharmacologycited by 0
  3. 2026SciR: A Controllable Benchmark for Scientific Reasoning in LLMscited by 0
  4. 2026Deceptive Grounding: Entity Attribution Failure in Clinical Retrieval-Augmented Generationcited by 0
  5. 2026Who Gets a Token, and What Does It Carry? Unequal Name Support and Concept Access in Large Language Modelscited by 0