SOURCE-LINKED INTELLIGENCE

MTDiag: A Multi-Turn Diagnostic Dataset Towards Clinically Meaningful LLM Evaluation

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Clinical diagnosis is fundamentally interactive and incremental, yet the dominant paradigm for evaluating Large Language Models (LLMs) in medicine remains static QA benchmarks or template-based dialogues. These benchmarks say little about whether a model can serve as a diagnostic agent in a dynamic clinical encounter, with LLMs showing significant accuracy and reliability degradation in multi-turn settings. To address this issue, we present MTDiag, a large multi-turn diagnostic dialogue dataset constructed from three heterogeneous sources: DDXPlus, MIMIC-IV, and published case reports (AJCR),

Read original source ↗ Open in workspace

recordType

paper

region

Global