AI RESEARCH
What Really Controls Temporal Reasoning in Large Language Models: Tokenisation or Representation of Time?
arXiv CS.AI
•
ArXi:2603.19017v1 Announce Type: cross We present MultiTempBench, a multilingual temporal reasoning benchmark spanning three tasks, date arithmetic, time zone conversion, and temporal relation extraction across five languages (English, German, Chinese, Arabic, and Hausa) and multiple calendar conventions (Gregorian, Hijri, and Chinese Lunar). MultiTempBench contains $15,000$ examples built by translating $750$ curated English questions and expanding each into controlled date-format variants. We evaluate 20 LLMs and.