Paper Accepted to ACL 2024 (Update: and Received the Area Chair Award!)

Our paper titled Distributional Inclusion Hypothesis and Quantifications: Probing for Hypernymy in Functional Distributional Semantics is accepted to ACL 2024, and received the Area Chair Award for the Semantics track! Getting the paper accepted to ACL 2024 was already a bonus, and I am truly grateful that this work is recognized among other brilliant papers out there. The camera-ready version is now available at the ACL Anthology.

Together with Functional Distributional Semantics at Scale, they complete a fuller picture of FDS, demonstrating how functional representations of words are used and learnt from the distributional information of corpora, and under what conditions are the representations learnt purely from corpora faithful via probing if hypernymy is respected. Establishing such close connections between the DIH, quantifications, and FDS is intriguing, and was not anticipated when I first started working on it. Nevertheless, it still bugs me that such analyses can only go so far to sentences with only one quantifier on a simple taxonomic hierarchy of nouns!