XWinograd

XWinograd is a multilingual Winograd Schema Challenge covering English, French, Japanese, Portuguese, Russian and Chinese.

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	reasoning
Subcategory	multilingual commonsense coreference resolution
Page status	active
Metric	accuracy
Direction	higher_is_better
Unit	percent
Dataset size	4449
Publisher	Yandex Research

What it measures

XWinograd tests commonsense pronoun/coreference resolution across six languages. Each item is a sentence with an ambiguous pronoun and two candidate referents, and the model must pick the referent that world knowledge and context make correct, resolving the same kind of ambiguity the original English Winograd Schema Challenge was built to test.

Task format

Binary choice; given a sentence with an ambiguous pronoun, pick which of two candidate referents it refers to.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub