Test of Time: A Benchmark for Evaluating LLMs on Temporal Reasoning

Our summary

A synthetic benchmark that isolates temporal reasoning from memorized facts, with a dedicated arithmetic split over time points and durations. Frontier models struggle on the calculation-heavy temporal tasks.

Why it matters

Shows date and duration arithmetic is a distinct, testable weakness even when the model isn't leaning on stale world knowledge.

Related findings (1)

• Date and duration arithmetic errorsLow

  Models miscompute differences between dates, weekdays, and durations across boundaries like months and leap years.

Cite this