Measuring Iterative Temporal Reasoning with Time Puzzles
arXiv:2601.07148v4 Announce Type: replace-cross Abstract: Tool use, such as web search, has become a standard capability even in freely available large language models (LLMs). However, existing benchmarks evaluate temporal reasoning mainly in…
Read the full story at arXiv cs.AI ↗
ImpactNotable 31/100
Why it mattersRule-based estimate: event keywords (+4); trust 6/10.
RegionsGlobal
Published4 h ago (Fri, 02 Oct 2026 04:00:00 GMT)
RetrievedFri, 02 Oct 2026 07:00:28 GMT via rss
ClassifiedFri, 02 Oct 2026 07:00:40 GMT by heuristic
AuthorZhengxiang Wang, Zeyu Dong