# Perspectives on neural proof nets

arXiv: https://arxiv.org/abs/2211.04141 · HTML: https://mlsanity.com/paper/2211.04141/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 30 references that are in the map.
- First posted on arXiv 8 Nov 2022.

## Builds on (most cited first, 5 of 5)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Sequence-to-Action: End-to-End Semantic Graph Generation for Semantic Parsing](https://mlsanity.com/paper/1809.00773/) (2018)
- [Neural Proof Nets](https://mlsanity.com/paper/2009.12702/) (2020)
- [AE THEL: Automatically Extracted Typelogical Derivations for Dutch](https://mlsanity.com/paper/1912.12635/) (2019)
- [Geometry-Aware Supertagging with Heterogeneous Dynamic Convolutions](https://mlsanity.com/paper/2203.12235/) (2022)

## Abstract (arXiv)

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.