CryptanalysisBench: Can LLMs Do Cryptanalysis?

CryptanalysisBench is an arXiv paper centered on whether large language models can perform cryptanalysis, apparently through a benchmark or evaluation suite. The available record contains only the title, URL, publication timestamp, and a brief Hacker News listing; it does not provide the abstract, benchmark composition, evaluated models, experimental protocol, or results. Those missing details prevent a reliable assessment of the paper’s technical contribution or any claims about LLM capability.