# Do Large Language Models have Shared Weaknesses in Medical Question Answering?

arXiv: https://arxiv.org/abs/2310.07225 · HTML: https://mlsanity.com/paper/2310.07225/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 39 references that are in the map.
- First posted on arXiv 11 Oct 2023.

## Builds on (most cited first, 26 of 26)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Large Language Models as Optimizers](https://mlsanity.com/paper/2309.03409/) (2023)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Capabilities of GPT-4 on Medical Challenge Problems](https://mlsanity.com/paper/2303.13375/) (2023)
- [Augmented Language Models: a Survey](https://mlsanity.com/paper/2302.07842/) (2023)
- [Towards Generalist Biomedical AI](https://mlsanity.com/paper/2307.14334/) (2023)
- [Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questions](https://mlsanity.com/paper/2308.11483/) (2023)
- [Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets](https://mlsanity.com/paper/2106.10328/) (2021)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [emrQA: A Large Corpus for Question Answering on Electronic Medical Records](https://mlsanity.com/paper/1809.00732/) (2018)
- [Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Dataset](https://mlsanity.com/paper/2306.03030/) (2023)
- [Energy Usage Reports: Environmental awareness as part of algorithmic accountability](https://mlsanity.com/paper/1911.08354/) (2019)

## Built on by (most cited first, 2 of 2)

- [Polish-English medical knowledge transfer: A new benchmark and results](https://mlsanity.com/paper/2412.00559/) (2024)
- [KNIGHT: Knowledge Graph-Driven Multiple-Choice Question Generation with Adaptive Hardness Calibration](https://mlsanity.com/paper/2602.20135/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have made rapid improvement on medical benchmarks, but their unreliability remains a persistent challenge for safe real-world uses. To design for the use LLMs as a category, rather than for specific models, requires developing an understanding of shared strengths and weaknesses which appear across models. To address this challenge, we benchmark a range of top LLMs and identify consistent patterns across models. We test 16 well-known LLMs on 874 newly collected questions from Polish medical licensing exams. For each question, we score each model on the top-1 accuracy and the distribution of probabilities assigned. We then compare these results with factors such as question difficulty for humans, question length, and the scores of the other models. LLM accuracies were positively correlated pairwise (0.39 to 0.58). Model performance was also correlated with human performance (0.09 to 0.13), but negatively correlated to the difference between the question-level accuracy of top-scoring and bottom-scoring humans (-0.09 to -0.14). The top output probability and question length were positive and negative predictors of accuracy respectively (p< 0.05). The top scoring LLM, GPT-4o Turbo, scored 84%, with Claude Opus, Gemini 1.5 Pro and Llama 3/3.1 between 74% and 79%. We found evidence of similarities between models in which questions they answer correctly, as well as similarities with human test takers. Larger models typically performed better, but differences in training, architecture, and data were also highly impactful. Model accuracy was positively correlated with confidence, but negatively correlated with question length. We find similar results with older models, and argue that these patterns are likely to persist across future models using similar training methods.