Leveraging Codebook Knowledge with NLI and ChatGPT for Zero-Shot Political Relation Classification

TL;DR

This paper explores zero-shot political relation classification using expert knowledge, leveraging ChatGPT and a novel NLI-based model called ZSP, demonstrating ZSP's superior performance and interpretability over traditional methods.

Contribution

Introduces ZSP, a new NLI-based model that improves zero-shot political relation classification by decomposing tasks and leveraging codebook knowledge, outperforming dictionary-based and some supervised methods.

Findings

ZSP outperforms dictionary-based methods.

ChatGPT shows strengths and limitations in zero-shot classification.

ZSP is competitive with supervised models.

Abstract

Is it possible accurately classify political relations within evolving event ontologies without extensive annotations? This study investigates zero-shot learning methods that use expert knowledge from existing annotation codebook, and evaluates the performance of advanced ChatGPT (GPT-3.5/4) and a natural language inference (NLI)-based model called ZSP. ChatGPT uses codebook's labeled summaries as prompts, whereas ZSP breaks down the classification task into context, event mode, and class disambiguation to refine task-specific hypotheses. This decomposition enhances interpretability, efficiency, and adaptability to schema changes. The experiments reveal ChatGPT's strengths and limitations, and crucially show ZSP's outperformance of dictionary-based methods and its competitive edge over some supervised models. These findings affirm the value of ZSP for validating event records and advancing ontology development. Our study underscores the efficacy of leveraging transfer learning and existing domain expertise to enhance research efficiency and scalability.