Montague Grammar Induction

2021/03/02 by Gene Louis Kim, Aaron Steven White · 1 voice
Computer Science · Arts and Humanities · #Natural Language Processing Techniques #Topic Modeling #Syntax, Semantics, Linguistic Variation

paper · pdf · doi:10.3765/salt.v30i0.4816

openalex publication_date 2021/03/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We propose a computational model for inducing full-fledged combinatory categorial grammars from behavioral data. This model contrasts with prior computational models of selection in representing syntactic and semantic types as structured (rather than atomic) objects, enabling direct interpretation of the modeling results relative to standard formal frameworks. We investigate the grammar our model induces when fit to a lexicon-scale acceptability judgment dataset – Mega Acceptability – focusing in particular on the types our model assigns to clausal complements and the predicates that select them.

Citations

• Learning to Map Sentences to Logical Form: Structured Classification with Probabilistic Categorial Grammars
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive\n Autoencoders
• Suspending OpenMP Tasks on Asynchronous Events: Extending the Taskwait Construct

Discussions

• For each subtask, I specify an interface, implement a good portion of the plumbing, and then give them a metric they need to optimize some component for–e.g. encoder-decoder networks that implement pa [bsky, 0 points, 1 comments]