Do Supervised Distributional Methods Really Learn Lexical Inference Relations?
Omer Levy, Steffen Remus, Chris Biemann, and Ido Dagan. Short paper in NAACL 2015. [pdf] [slides]
Distributional representations of words have been recently used in supervised settings for recognizing lexical inference relations between word pairs, such as hypernymy and entailment. We investigate a collection of these state-of-the-art methods, and show that they do not actually learn a relation between two words. Instead, they learn an independent property of a single word in the pair: whether that word is a “prototypical hypernym”.
Data
The datasets used in this paper, as well as their train/test splits, are available for download here.
.
.
.
.
.
Omer Levy 