SuperGLUE CB (CommitmentBank)

SuperGLUE's three-class entailment recast of CommitmentBank: 250/56/250 English pairs, scored with accuracy and macro-F1.

Also known as: CB, CommitmentBank, SuperGLUE_CB

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.
Categoryreasoning
Subcategorythree-class textual entailment from speaker commitment (SuperGLUE)
Page statussaturated
Metricaccuracy and macro-F1 (unweighted mean of per-class F1); SuperGLUE reports both
Directionhigher_is_better
Unit%
Dataset size556
Dataset licenceother
PublisherNew York University (SuperGLUE); original CommitmentBank from Ohio State / Carnegie Mellon / Ohio State authors

What it measures

SuperGLUE CB asks whether a premise commits its author to a hypothesis extracted from an embedded clause. The original CommitmentBank labels how committed a speaker is to that clause on a Likert scale. SuperGLUE recasts the items as three-class textual entailment: entailment, contradiction, or neutral. Premises come from the Wall Street Journal, British National Corpus fiction, and Switchboard. SuperGLUE keeps a subset with inter-annotator agreement above 80% and uses its own split. The task is English text, single-turn classification, not open-ended generation.

Task format

Three-way classification: premise plus hypothesis to entailment, contradiction, or neutral; English.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub