ArticleScientific reports2025
Limitations of large language models in clinical problem-solving arising from inflexible reasoning.
Article in Scientific reports, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 40 papers, 1 of them a synthesis that pooled it.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
40 citing papers in PubMed, 1 synthesis or guideline pooled it.
- The impact of generative artificial intelligence on clinical skills and knowledge acquisition in medical undergraduates: a systematic review and meta-analysis.BMC medical education · 2026Pooled it
- Evaluation of DeepSeek with a RAG-based knowledge base in preoperative assessment for arthroplasty to improve the decision-making accuracy of junior physicians.Arthroplasty (London, England) · 2026Article
- Effect of large language model on diagnostic accuracy and clinical completeness among nephrology fellows managing transplant infection.International urology and nephrology · 2026Article
- Concordance between GPT-4 and a multidisciplinary tumor board in pancreatic cancer: A prospective pilot study.Langenbeck's archives of surgery · 2026Observational
- Performance, Failures, and Oversight of a Large Language Model Agent for Clinical Data Analysis: Evaluation Study.Journal of medical Internet research · 2026Article
- Role prompting modulates linguistic style but not clinical decision structure in GPT-5 tumour board simulation.NPJ digital medicine · 2026Article
- Stepwise Diagnostic Evaluation of Chinese Large Language Models: Comparative Study of Common and Rare Diseases.Journal of medical Internet research · 2026Article
- Article
- Performance of Large Language Models in Oral Health Consultations and the Consistency of the 'AI-as-a-Judge' Framework.International dental journal · 2026Article
- Performance of multimodal large language models on image-based surgical anatomy, anatomical pathology, and radiology questions.Anatomical sciences education · 2026Article
- Integrating Genetic, Environmental, Cognitive, and Temperament Data for ADHD Prediction in Explainable Deep Learning Models.medRxiv : the preprint server for health sciences · 2026Article
- Can artificial intelligence diagnose the invisible? A scenario-based evaluation of AI performance in cracked tooth diagnosis.Odontology · 2026Article
- WiNGPT-32B: An Open-Source, Locally Deployable LLM for RECIST Assessment via Chained Task Execution Using Radiology Report Text.Diagnostics (Basel, Switzerland) · 2026Article
- Enhancing Physician Resilience to Generative AI: Multilevel Framework for Shared Authority, Verification, and Skill Preservation.Journal of medical Internet research · 2026Article
- Benchmarking reliability and calibration of LLMs for multi-cancer early detection test communication.JAMIA open · 2026Article
- Evaluating the performance of ChatGPT-5, Claude, and DeepSeek for statistical analysis in medicine.Internal and emergency medicine · 2026Article
- Performance comparison of a neuro-symbolic large language model system versus conventional AI models and human experts in cholangitis management.BMC medical informatics and decision making · 2026Article
- Clinical Plausibility in Large Language Model Robustness Testing for Medicine: A Scoping Review.Journal of medical systems · 2026Article
- Evaluation of large language models and retrieval-augmented generation for clinical reasoning in pediatric myopia: a 50-case real-world study.Scientific reports · 2026Article
- All Models are Wrong, Some are Annotated: Automating Metadata in Biomedical Repositories.bioRxiv : the preprint server for biology · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
6 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Large Language Models (LLMs) have attained human-level accuracy on medical question-answer (QA) benchmarks. However, their limitations in navigating clinical scenarios requiring flexible reasoning have recently been shown, raising concerns about the robustness and generalizability of LLM reasoning across diverse, real-world medical tasks. To probe potential LLM failure modes in clinical problem-solving, we present the medical abstraction and reasoning corpus (mARC-QA). mARC-QA assesses clinical reasoning through scenarios designed to exploit the Einstellung effect-the fixation of thought arising from prior experience, targeting LLM inductive biases toward inflexible pattern matching from their training data rather than engaging in flexible reasoning. We find that LLMs, including current state-of-the-art o1, Gemini, Claude, and DeepSeek models, perform poorly compared to physicians on mARC-QA, often demonstrating lack of commonsense medical reasoning and a propensity to hallucinate. In addition, uncertainty estimation analyses indicate that LLMs exhibit overconfidence in their answers, despite their limited accuracy. The failure modes revealed by mARC-QA in LLM medical reasoning underscore the need to exercise caution when deploying these models in clinical settings.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.