Multilingual Relation Classification via Efficient and Effective Prompting

TL;DR

This paper introduces a prompt-based approach for multilingual relation classification that is efficient, effective, and performs well across various data regimes and languages, requiring minimal translation effort.

Contribution

It is the first to apply prompt-based methods to multilingual relation classification, demonstrating strong performance with minimal translation and language-specific knowledge.

Findings

Outperforms fine-tuning XLM-R_EM and null prompts in supervised and few-shot settings.

Significantly outperforms random baseline in zero-shot scenarios.

Effective across 14 languages and various prompt variants.

Abstract

Prompting pre-trained language models has achieved impressive performance on various NLP tasks, especially in low data regimes. Despite the success of prompting in monolingual settings, applying prompt-based methods in multilingual scenarios has been limited to a narrow set of tasks, due to the high cost of handcrafting multilingual prompts. In this paper, we present the first work on prompt-based multilingual relation classification (RC), by introducing an efficient and effective method that constructs prompts from relation triples and involves only minimal translation for the class labels. We evaluate its performance in fully supervised, few-shot and zero-shot scenarios, and analyze its effectiveness across 14 languages, prompt variants, and English-task training in cross-lingual settings. We find that in both fully supervised and few-shot scenarios, our prompt method beats competitive baselines: fine-tuning XLM-R_EM and null prompts. It also outperforms the random baseline by a large margin in zero-shot experiments. Our method requires little in-language knowledge and can be used as a strong baseline for similar multilingual classification tasks.