Perspectives on neural proof nets
This work addresses proof search for computational linguistics, but it appears incremental as it builds on existing graph neural network techniques.
The paper tackles the problem of proof search in type-logical grammars by introducing a novel method that splits the task into graph generation and vertex labeling using neural networks, contrasting with the standard supertagging approach.
In this paper I will present a novel way of combining proof net proof search with neural networks. It contrasts with the 'standard' approach which has been applied to proof search in type-logical grammars in various different forms. In the standard approach, we first transform words to formulas (supertagging) then match atomic formulas to obtain a proof. I will introduce an alternative way to split the task into two: first, we generate the graph structure in a way which guarantees it corresponds to a lambda-term, then we obtain the detailed structure using vertex labelling. Vertex labelling is a well-studied task in graph neural networks, and different ways of implementing graph generation using neural networks will be explored.