The reviewed record of science sign in
Pith

arxiv: 2402.12691 · v2 · pith:WJTETAOU · submitted 2024-02-20 · cs.CL

Tree-Planted Transformers: Unidirectional Transformer Language Models with Implicit Syntactic Supervision

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:WJTETAOUrecord.jsonopen to challenge →

classification cs.CL
keywords syntactictransformerstructuresslmsefficiencylanguageunidirectionalexplicit
0
0 comments X
read the original abstract

Syntactic Language Models (SLMs) can be trained efficiently to reach relatively high performance; however, they have trouble with inference efficiency due to the explicit generation of syntactic structures. In this paper, we propose a new method dubbed tree-planting: instead of explicitly generating syntactic structures, we "plant" trees into attention weights of unidirectional Transformer LMs to implicitly reflect syntactic structures of natural language. Specifically, unidirectional Transformer LMs trained with tree-planting will be called Tree-Planted Transformers (TPT), which inherit the training efficiency from SLMs without changing the inference efficiency of their underlying Transformer LMs. Targeted syntactic evaluations on the SyntaxGym benchmark demonstrated that TPTs, despite the lack of explicit generation of syntactic structures, significantly outperformed not only vanilla Transformer LMs but also various SLMs that generate hundreds of syntactic structures in parallel. This result suggests that TPTs can learn human-like syntactic knowledge as data-efficiently as SLMs while maintaining the modeling space of Transformer LMs unchanged.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.