LongCoT: Benchmarking Long-Horizon Chain-of-Thought Reasoning
University of Oxford · University of Maryland, College Park · Lawrence Livermore National Labs · KAUST · Lawrence Livermore National Laboratory · MATS · Hexo labs · Mohamed bin Zayed University of Artificial Intelligence · Lawrence Livermore National Laboratory (LLNL) · Oxford · University of Tübingen
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
As language models are increasingly deployed for complex autonomous tasks, their ability to reason accurately over longer horizons becomes critical. An essential component of this ability is planning and managing a long, complex chain-of-thought (CoT). We introduce LongCoT, a scalable benchmark of 2,500 expert-designed problems spanning chemistry, mathematics, computer science, chess, and logic to isolate and directly measure the long-horizon CoT reasoning capabilities of frontier models. Problems consist of a short input with a verifiable answer; solving them requires navigating a graph of interdependent steps that span tens to hundreds of thousands of reasoning tokens. Each local step is individually tractable for frontier models, so failures reflect long-horizon reasoning limitations. At release, the best models achieve <10% accuracy (GPT 5.2: 9.8%; Gemini 3 Pro: 6.1%) on LongCoT, revealing a substantial gap in current capabilities. Overall, LongCoT provides a rigorous measure of long-horizon reasoning, tracking the ability of frontier models to reason reliably over extended periods.