MICROSOFT | The KnowRef Coreference Corpus: a resource for training and evaluating common sense in AI
AI has made major strides in the last decade, from beating the world champion of Go, to learning how to program, to telling fantastical short stories. However, a basic human trait continues to elude machines: common sense.
McGill University researchers and students, notably Ali Emami, Paul Trichelair, and Jackie Chi Kit Cheung, collaborated with Microsoft researchers on a recent paper, “The KnowRef Coreference Corpus: Removing Gender and Number Cues for Difficult Pronominal Anaphora Resolution,” which attempts to answer this question. It will appear at the 2019 Annual Meeting of the Association for Computational Linguistics (ACL) in Florence, Italy. The paper introduces a new resource for training and evaluating common sense in machines, the KnowRef coreference corpus. This benchmark contains over 8,000 annotated text passages from the web that exhibit natural, knowledge-oriented instances of pronominal coreference.