Anachronisms

A 230-item BIG-bench Yes/No task that asks whether an English sentence contains a temporally impossible mix of people, objects, or dates.

Also known as: Identifying Anachronisms

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.

Category	knowledge
Subcategory	BIG-bench Yes/No prochronism detection (230 items)
Page status	unknown
Metric	multiple_choice_grade
Direction	higher_is_better
Unit	%
Dataset size	230
Dataset licence	Apache-2.0
Publisher	Google (BIG-bench collaboration)

What it measures

anachronisms presents one English sentence and asks whether it contains non-contemporaneous elements. The authors restricted items to prochronisms (temporally impossible pairings), not old-fashioned but possible uses. Half the sentences are anachronistic and half are not; some non-anachronisms are still factually false. The skill is temporal consistency, not trivia correctness. It is not [bbh](bbh.md), which does not include this task.

Task format

Two-option multiple choice (Yes/No) with preferred metric multiple_choice_grade. example_output_prefix asks whether the preceding sentence contains non-contemporaneous (anachronistic) elements. append_choices_to_input is false. Canary GUID embedded.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub