David Arps — accepted AI-conference papers

2024

Multilingual Nonce Dependency Treebanks: Understanding how Language Models Represent and Process Syntactic Structure

David Arps, Laura Kallmeyer, Younes Samih, Hassan Sajjad

NAACL 2024long

We introduce SPUD (Semantically Perturbed Universal Dependencies), a framework for creating nonce treebanks for the multilingual Universal Dependencies (UD) corpora. SPUD data satisfies syntactic argument structure, provides syntactic annotations, and ensures grammaticality via language-specific rul…

Cited by 1 Source ↗PDF ↗code ↗Scholar ↗

2022

Probing for Constituency Structure in Neural Language Models

David Arps, Younes Samih, Laura Kallmeyer, Hassan Sajjad

EMNLP 2022finding

In this paper, we investigate to which extent contextual neural language models (LMs) implicitly learn syntactic structure. More concretely, we focus on constituent structure as represented in the Penn Treebank (PTB). Using standard probing techniques based on diagnostic classifiers, we assess the a…

Cited by 33 Source ↗PDF ↗code ↗Scholar ↗