Neuro-symbolic Syntactic Parsing: Shaping a Neural Network with the CYK Algorithm

2026-05-29

AI Summary

This work proposes CYKNN, a novel neural architecture that directly embeds the Cocke–Younger–Kasami (CYK) context-free grammar parsing algorithm into a recurrent neural network. By formulating the CYK algorithm through differentiable matrix-vector operations, CYKNN enables end-to-end trainable encoding of symbolic parsing within a neural framework, thereby achieving a deep integration of symbolic reasoning and neural computation. Empirical results demonstrate that CYKNN substantially outperforms both large language models with over 20 billion parameters and LoRA-finetuned variants of the Qwen model family on simple grammatical tasks. This approach establishes a promising new direction for neuro-symbolic systems by combining the expressivity of formal grammars with the learning capabilities of neural networks.

Abstract

In this paper, we show the possibility of a direct injection of algorithms into neural network architecture. We focus on a complex algorithm, that is, Cocke-Youger-Kasami (CYK) for parsing context-free grammars in Chomsky Normal Form and we propose CYKNN, a simple recurrent neural network architecture for encoding the CYK algorithm in trainable matrix-vector multiplications.We experimented with a very simple grammar with 4 variations showing that our approach outperforms existing LLMs with more than 20B parameters with an in-context learning setting and smaller LLMs of the Qwen family fine-tuned with LoRA. Our attempt paves the way to a different approach to neuro-symbolic methodologies.

Problem

• neuro-symbolic
• syntactic parsing
• CYK algorithm
• context-free grammar
• neural architecture

Innovation

• neuro-symbolic
• CYK algorithm
• neural architecture
• grammar parsing
• in-context learning