🧬 Bad news for medical LLMs. This paper finds that top medical AI models often match patterns instead of truly reasoning. Small wording tweaks cut accuracy by up to 38% on validated questions. The team took 100 MedQA questions, replaced the correct choice with None of the other answers, then kept the 68 items where a clinician confirmed that switch as correct. If a model truly reasons, it should still reach the same clinical decision despite that label swap. They asked each model to explain its steps before answering and compared accuracy on the original versus modified items. All 6 models dropped on the NOTA set, the biggest hit was 38%, and even the reasoning models slipped. That pattern points to shortcut learning, the systems latch onto answer templates rather than working through the clinical logic. Overall, the results show that high benchmark scores can mask a robustness gap, because small format shifts expose shallow pattern use rather than clinical reasoning.

Interesting findings. Probably this evaluation should be done again with Gemini 2.5 Pro , GPT5 Pro, Claude 4.0 . The models they have used seem a little old now: "We evaluated 6 models spanning different architectures and capabilities: DeepSeek-R1 (model 1), o3-mini (reasoning models) (model 2), Claude-3.5 Sonnet (model 3), Gemini-2.0-Flash (model 4), GPT-4o (model 5), and Llama-3.3-70B (model 6)"