PreCo: A Large-scale Dataset in Preschool Vocabulary for Coreference Resolution
This dataset provides a resource for researchers in natural language processing to better analyze coreference resolution, though it is incremental as it builds on existing datasets like OntoNotes.
The authors introduced PreCo, a large-scale English dataset for coreference resolution designed to address challenges like low training-test overlap, using a corpus of about 38K documents and 12.4M words from preschool vocabulary, which improved error analysis efficiency compared to OntoNotes and enabled quantification of mention detector influence.
We introduce PreCo, a large-scale English dataset for coreference resolution. The dataset is designed to embody the core challenges in coreference, such as entity representation, by alleviating the challenge of low overlap between training and test sets and enabling separated analysis of mention detection and mention clustering. To strengthen the training-test overlap, we collect a large corpus of about 38K documents and 12.4M words which are mostly from the vocabulary of English-speaking preschoolers. Experiments show that with higher training-test overlap, error analysis on PreCo is more efficient than the one on OntoNotes, a popular existing dataset. Furthermore, we annotate singleton mentions making it possible for the first time to quantify the influence that a mention detector makes on coreference resolution performance. The dataset is freely available at https://preschool-lab.github.io/PreCo/.