CritPt
B待确认Benchmark垂直领域科学开放apache-2.0
发布方：CritPt-Benchmark
热度25.2▲ 0.1
下载量 · 30天
4,402
Hugging Face
GitHub Stars
—
代码仓库
论文被引
11
Semantic Scholar
跑分模型 · 30天
—
Leaderboard results
简介
Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark |🌐 Website | GitHub | 📖 Paper | Dataset description CritPt (Complex Research using Integrated Thinking – Physics Test; reads as "critical point") is the first benchmark designed to test LLMs on unpublished, research-level reasoning tasks that broadly covers modern physics research areas, including condensed matter, quantum physics, atomic, molecular & optical physics, astrophysics… See the full description on the dataset page: https://huggingface.co/datasets/CritPt-Benchmark/CritPt.