Microsoft's MAI-DxO hits 83% diagnostic accuracy on 304 NEJM cases vs 36% for physicians
jiqizhixin · x · 2026-11-03
Microsoft's Science paper on sequential diagnosis reports MAI-DxO achieving 83% diagnostic accuracy on 304 complex NEJM cases, versus 36% for physicians using reference resources in a controlled evaluation.
- SDBench benchmark: a gatekeeper reveals only a short case summary, then releases evidence as questions and test requests arrive; it scores both diagnosis accuracy and estimated testing cost
- The Diagnostic Orchestrator coordinates complementary reasoning roles to update candidate diagnoses, challenge assumptions, and weigh the value of further tests; some unrecorded test results are simulated
- Caveat: retrospective, text-based case evaluations—not a demonstration of independent clinical practice; physician comparison is individuals, not teams
More from Models
- Rumor: OpenAI model solved Navier-Stokes millennium problem and 300+ open math problems at ~3h compute each — haider1 · 2026-11-03
- AgentTime Update: GPT Models Lead Time-Awareness, Best Forecast Miss Is 2.26× — maksym_andr · 2026-11-03
- paul_cal catalogs the untold failures of LLM tuning — paul_cal · 2026-11-03
- DeepSeek V5 reportedly to be pretrained on Huawei Ascend chips, Ascend 960DT due Q1 2027 — teortaxesTex · 2026-11-03
- Rumors swirl of massive GPT-6.5 pretraining run reportedly using millions of GPUs — VraserX · 2026-11-03
- Tsinghua spinoff Apex Intelligence open-weights Apex-35B, built by automated AI research system — jiqizhixin · 2026-11-03