Common Morpheme

A 50-item BIG-bench 4-way task (25 English, 25 French) that asks for the shared morpheme meaning across a word list.

Also known as: BIG-bench common_morpheme

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 4-way shared-morpheme meaning (25 English + 25 French)
Page status	unknown
Metric	multiple_choice_grade
Direction	higher_is_better
Unit	%
Dataset size	50
Dataset licence	Apache-2.0
Publisher	Google (BIG-bench collaboration); task author at University of Toronto

What it measures

common_morpheme lists four words that share a derivational morpheme and asks which of four glosses is that shared meaning. Shammie (Shyamolima) Debnath wrote 25 English and 25 French items. Distractors relate to one listed word but not to the common piece. The probe is morpheme-level semantics, not inflectional agreement.

Task format

Four-way multiple choice with preferred metric multiple_choice_grade. append_choices_to_input is true. Canary GUID embedded.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub