ANVAN-LS: Lexical Substitution for Evaluating Compositional Distributional Models
ANVAN-LS is a lexical substitution dataset for CDSM evaluation sampled from an English-language corpus with manual “all-words” lexical substitution annotation. The sentences all have the traditional Adjective-Noun-Verb-Adjective-Noun (ANVAN) format.
Maja Buljan, Sebastian Padó, Jan Snajder. Lexical Substitution for Evaluating Compositional Distributional Models. Proceedings of NAACL 2018, New Orleans, LA.