arXiv cs.AI· Pengkai Wang, Wei-Wei Zhang, Yan Li, Min Tang, Zhitian Hou, Zeyu Liu, Guanghao Zhu, Yuanyi Wang, Yanggan Gu, Wenjun Wang, Minheng Ni, Congkai Xie, Zhijie Sang, Jianmin Wu, Ying Sun, Hongxia Yang·· 4 小时前AI 评分33
NPCBench：面向鼻咽癌多模态诊疗路径推理的临床学徒式评测基准
NPCBench: A Clinical Apprenticeship Benchmark for Multimodal Care-Pathway Reasoning in Nasopharyngeal Carcinoma
AI 导读
NPCBench 是一个针对鼻咽癌（NPC）的临床学徒式评测基准，涵盖指南知识、病例规划、纵向复查、CarePath 整合与开放式问诊，被 NeurIPS 2026 评测与数据集赛道接收。
正文 · 原文
Authors:Pengkai Wang, Wei-Wei Zhang, Yan Li, Min Tang, Zhitian Hou, Zeyu Liu, Guanghao Zhu, Yuanyi Wang, Yanggan Gu, Wenjun Wang, Minheng Ni, Congkai Xie, Zhijie Sang, Jianmin Wu, Ying Sun, Hongxia Yang
Abstract:Current medical benchmarks typically break down clinical competence into isolated factual questions, single-image tasks, or static case vignettes. These tests score individual answers, but reveal less about whether a model can integrate subspecialty knowledge with multimodal evidence and track a patient's condition throughout a course of care. To close this gap, we introduce NPCBench, a clinical apprenticeship benchmark for nasopharyngeal carcinoma (NPC). NPCBench mirrors specialist training, from guideline and foundational knowledge to case planning, longitudinal review, CarePath integration, and open-ended consultation. It contains 4,777 component and consultation items and a 100-patient CarePath cohort with 417 scored diagnostic and care tasks. The episode-level evaluation tests whether diagnostic and care answers are jointly correct within a care pathway. By linking stepwise tasks to patient-level care trajectories, NPCBench evaluates whether models ground decisions in visual evidence, interpret changes in patient state, and meet clinical requirements in consultation. Across 30 evaluated models, we observe a composition gap: strong performance on individual tasks does not ensure jointly correct diagnostic and care answers for the same patient. Among 21 multimodal models, strict CarePath success reaches at most 23%. Paired controls and physician calibration further examine evidence grounding, longitudinal revision, and consultation judging. Overall, NPCBench broadens the evaluation of medical artificial intelligence beyond static question-answering paradigms to an apprenticeship-oriented assessment of clinical workflows, thereby offering a testbed for evidence-based and temporally coherent reasoning in oncology care.
| Comments: | Accepted at the 40th Conference on Neural Information Processing Systems (NeurIPS 2026), Evaluations and Datasets Track |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.08369 [cs.AI] |
| (or arXiv:2611.08369v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08369 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wang Pengkai [view email]
[v1]
Thu, 5 Nov 2026 17:48:55 UTC (3,100 KB)
来源：arXiv cs.AI · arxiv.org