CritPt-Benchmark/CritPt

CritPt（复杂研究综合思维-物理测试，读作“临界点”）是首个设计用于测试大型语言模型在未发表的研究级别推理任务上的基准数据集。该数据集广泛涵盖了现代物理研究的各个领域，包括凝聚态物理、量子物理、原子物理、分子与光学物理、天体物理、统计物理、核物理、高能物理、数学物理、流体动力学、非线性动力学和生物物理。该数据集由全球30多个顶尖机构的50多名活跃物理研究人员制作，其中包括资深博士生、博士后和教授，共包含71个挑战和190个检查点。

CritPt (Complex Research using Integrated Thinking – Physics Test; reads as "critical point") is the first benchmark designed to test large language models on unpublished, research-level reasoning tasks that broadly covers modern physics research areas, including condensed matter, quantum physics, atomic, molecular & optical physics, astrophysics, statistical physics, nuclear physics, high energy physics, mathematical physics, fluid dynamics, nonlinear dynamics and biophysics. The dataset is crafted by a team of 50+ active physics researchers from 30+ leading institutions worldwide, including senior Ph.D. Candidates, Postdocs, and Professors, containing 71 challenges and 190 checkpoints.