paper-with-me

홈 › Papers

Unraveling Syntax: Language Modeling and the Substructure of Grammars

2025-10-02 · Laura Ying Schulz, Daniel Mitropolsky, Tomaso Poggio arxiv

While language models achieve impressive results, their learning dynamics are far from understood. Many domains of interest -- such as natural language syntax, coding languages, arithmetic -- are captured by context-free grammars (CFGs). In this work, we extend prior work on neural language modeling of CFGs in a novel direction: how language modeling behaves with respect to CFG substructure, namely subgrammars. We define subgrammars, and prove a set of fundamental theorems connecting language modeling and subgrammars. We show that language modeling loss recurses linearly over its top-level subgrammars; applied recursively, the loss decomposes into losses for "irreducible" subgrammars. Under additional assumptions, and empirically, parametrized models learn subgrammars in parallel, unlike children who first master simple substructures. We find that subgrammar pretraining can improve final performance, but only for tiny models relative to the grammar, while alignment analyses show that pretraining consistently leads to internal representations that better reflect the grammar's substructure.

📄 PDF Abstract BibTeX arXiv:2510.02524

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Unsupervised Recurrent Neural Network Grammars

2019-04-07 · NAACL 2019 6 · Yoon Kim, Alexander M. Rush, Lei Yu, Adhiguna Kuncoro 외

Recurrent neural network grammars (RNNG) are generative models of language which jointly model syntax and surface structure by incrementally generating a syntax tree and sentence in a top-down, left-to-right order. Super…

Constituency Grammar InductionLanguage ModelingLanguage ModellingSentence+1

What Do Recurrent Neural Network Grammars Learn About Syntax?

2016-11-17 · EACL 2017 4 · Adhiguna Kuncoro, Miguel Ballesteros, Lingpeng Kong, Chris Dyer 외

Recurrent neural network grammars (RNNG) are a recently proposed probabilistic generative modeling family for natural language. They show state-of-the-art language modeling and parsing performance. We investigate what in…

Constituency ParsingDependency ParsingLanguage ModelingLanguage Modelling

Probabilistic modeling of the syntax semantic interface using probabilistic context free grammars, Experiments with FrameNet (Mod\'elisation probabiliste de l'interface syntaxe s\'emantique \`a l'aide de grammaires hors contexte probabilistes Exp\'eriences avec FrameNet) [in French]

2014-07-01 · JEPTALNRECITAL 2014 7 · Olivier Michalon
Semantic Parsing

Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale

2022-03-01 · Laurent Sartran, Samuel Barrett, Adhiguna Kuncoro, Miloš Stanojević 외

We introduce Transformer Grammars (TGs), a novel class of Transformer language models that combine (i) the expressive power, scalability, and strong performance of Transformers and (ii) recursive syntactic compositions, …

Inductive BiasLanguage ModelingLanguage ModellingSentence

Dependency Transformer Grammars: Integrating Dependency Structures into Transformer Language Models

2024-07-24 · Yida Zhao, Chao Lou, Kewei Tu

Syntactic Transformer language models aim to achieve better generalization through simultaneously modeling syntax trees and sentences. While prior work has been focusing on adding constituency-based structures to Transfo…

ARCInductive BiasLanguage ModelingLanguage Modelling