ResearchBenchmarksMedical Diagnosis on MD DXFollow10.5Worst Case Interaction LengthGoT10.24811.94913.6515.351Feb 2, 2026Evaluation ResultsMethodMethodLinksWorst Case Interaction LengthGoTBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0210.5UoTBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0212GoTBackbone=GPT 4.1Backbone=GPT 4.12026.0212.2DPBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0212.4UoTBackbone=GPT 4.1Backbone=GPT 4.12026.0213DCBackbone=Qwen 2.5 72B...Backbone=Qwen 2.5 72B Instruct2026.0213.3DCBackbone=GPT 4.1Backbone=GPT 4.12026.0216.2DPBackbone=GPT 4.1Backbone=GPT 4.12026.0216.8