# Constructive Type-Logical Supertagging with Self-Attention Networks

## TL;DR

This paper introduces a self-attention based supertagger for type-logical grammar induction, capable of learning syntax and semantics, and generalizing to unseen complex types, surpassing previous limitations.

## Contribution

It presents a novel attention-based supertagger that learns grammar types inductively, improving generalization and handling unseen complex types in type-logical grammar.

## Key findings

- Achieves high overall type accuracy
- Learns syntax and semantics of the grammar
- Constructs complex types never seen during training

## Abstract

We propose a novel application of self-attention networks towards grammar induction. We present an attention-based supertagger for a refined type-logical grammar, trained on constructing types inductively. In addition to achieving a high overall type accuracy, our model is able to learn the syntax of the grammar's type system along with its denotational semantics. This lifts the closed world assumption commonly made by lexicalized grammar supertaggers, greatly enhancing its generalization potential. This is evidenced both by its adequate accuracy over sparse word types and its ability to correctly construct complex types never seen during training, which, to the best of our knowledge, was as of yet unaccomplished.