On Eliciting Syntax from Language Models via Hashing

TL;DR

This paper presents a novel unsupervised method leveraging binary representations in language models to induce high-quality syntactic parsing trees efficiently, without supervised data.

Contribution

It introduces a first-order bit-level CKY, a contrastive hashing framework, and a new loss function for unsupervised syntax induction from pre-trained models.

Findings

Achieves competitive parsing performance on multiple datasets.

Demonstrates effectiveness of binary representations for syntax extraction.

Provides an efficient, low-cost approach to grammar induction.

Abstract

Unsupervised parsing, also known as grammar induction, aims to infer syntactic structure from raw text. Recently, binary representation has exhibited remarkable information-preserving capabilities at both lexicon and syntax levels. In this paper, we explore the possibility of leveraging this capability to deduce parsing trees from raw text, relying solely on the implicitly induced grammars within models. To achieve this, we upgrade the bit-level CKY from zero-order to first-order to encode the lexicon and syntax in a unified binary representation space, switch training from supervised to unsupervised under the contrastive hashing framework, and introduce a novel loss function to impose stronger yet balanced alignment signals. Our model shows competitive performance on various datasets, therefore, we claim that our method is effective and efficient enough to acquire high-quality parsing trees from pre-trained language models at a low cost.