A Scalable Tool for Measuring Manner and Result Verbs in Developmental Language Research

TL;DR

This paper introduces a scalable computational tool that uses large language models and a RoBERTa classifier to identify manner and result verbs in sentences, aiding developmental language research.

Contribution

It extends annotated verb classes using language models and trains a classifier achieving up to 89.6% accuracy for verb classification tasks.

Findings

Achieved up to 89.6% accuracy in classifying manner and result verbs.

Extended coverage from VerbNet to 436 classes using language models.

Provided a scalable tool for future developmental language studies.

Abstract

Manner and result verbs encode different aspects of event structure and have been discussed in developmental work as a potentially informative distinction for studying early verb learning. However, this distinction remains difficult to measure at scale because large annotated resources for manner and result classification are not currently available. We present a computational approach for identifying manner and result verbs in sentence context. Using linguistically informed prompts, we generate sentence-level annotations with large language models over data drawn from MASC and InterCorp, extending coverage from previously annotated portions of VerbNet to 436 classes. We then train a RoBERTa-based classifier on these annotations and evaluate it on three held-out gold-standard datasets, including previously annotated items and a new expert-annotated set. Across these evaluations, the model shows promising performance, with average accuracy up to 89.6%. We present this work as a scalable measurement tool that can support future research on verb semantics in developmental and other language datasets, while noting that further validation is needed for borderline cases, mixed manner/result verbs, and downstream developmental applications.