Small Models Are (Still) Effective Cross-Domain Argument Extractors
This work addresses the challenge of cross-domain argument extraction for NLP researchers, offering an incremental improvement by demonstrating the efficacy of smaller models over large language models.
The study tackled the problem of effective ontology transfer in event argument extraction by comparing question answering and template infilling techniques across six datasets, showing that smaller models trained on a source ontology can outperform GPT-3.5 and GPT-4 in zero-shot performance.
Effective ontology transfer has been a major goal of recent work on event argument extraction (EAE). Two methods in particular -- question answering (QA) and template infilling (TI) -- have emerged as promising approaches to this problem. However, detailed explorations of these techniques' ability to actually enable this transfer are lacking. In this work, we provide such a study, exploring zero-shot transfer using both techniques on six major EAE datasets at both the sentence and document levels. Further, we challenge the growing reliance on LLMs for zero-shot extraction, showing that vastly smaller models trained on an appropriate source ontology can yield zero-shot performance superior to that of GPT-3.5 or GPT-4.