Pith. sign in

REVIEW 1 cited by

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

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2203.00633 v2 pith:ULQIJHMT submitted 2022-03-01 cs.CL

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

classification cs.CL
keywords languagesyntactictransformermodelingmodelsfindgrammarsperplexity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

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, which here are implemented through a special attention mask and deterministic transformation of the linearized tree. We find that TGs outperform various strong baselines on sentence-level language modeling perplexity, as well as on multiple syntax-sensitive language modeling evaluation metrics. Additionally, we find that the recursive syntactic composition bottleneck which represents each sentence as a single vector harms perplexity on document-level language modeling, providing evidence that a different kind of memory mechanism -- one that is independent of composed syntactic representations -- plays an important role in current successful models of long text.

discussion (0)

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

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling

    cs.CL 2021-12 unverdicted novelty 6.0

    Semantic constituency graphs outperform syntactic constituency and dependency structures from seven formalisms when added to a Transformer for language modeling.