Known Unknowns

46-item BIG-bench Lite set that pairs a known fact with a question whose honest answer is Unknown.

Also known as: Known Unknowns (Hallucinations)

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.
Categoryknowledge
SubcategoryBIG-bench Lite Yes/No pairs that prefer Unknown over a fabricated fact (46 items)
Page statusunknown
Metricmultiple_choice_grade
Directionhigher_is_better
Unit%
Dataset size46
Dataset licenceApache-2.0
PublisherGoogle (BIG-bench collaboration)

What it measures

known_unknowns tests whether a model assigns higher probability to Unknown than to a specific but unfounded fact when the true answer is not knowable. Each unknown item is paired with a similar known question so a few-shot run cannot win by always saying Unknown. Imagined answers were often taken from davinci generations. The task is one narrow hallucination probe, not free-form fact checking. It is a BIG-bench Lite task and is not in [bbh](bbh.md).

Task format

Two-option multiple choice with preferred metric multiple_choice_grade and append_choices_to_input true. Dummy-model header: 46 multiple-choice queries. Canary GUID embedded. Gold is balanced 23 Unknown / 23 known answers.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub