X-PuDu at SemEval-2022 Task 6: Multilingual Learning for English and Arabic Sarcasm Detection

TL;DR

This paper presents a multilingual sarcasm detection system using fine-tuned pre-trained language models, achieving top rankings in SemEval-2022 tasks for English and Arabic sarcasm identification.

Contribution

The novel approach fine-tunes ERNIE-M and DeBERTa models for multilingual sarcasm detection in English and Arabic, demonstrating competitive performance.

Findings

Ranked second in Task A for one-sentence detection

Achieved first in Task C for sentence-pair detection in English

Performed well across multiple sarcasm detection tasks

Abstract

Detecting sarcasm and verbal irony from people's subjective statements is crucial to understanding their intended meanings and real sentiments and positions in social scenarios. This paper describes the X-PuDu system that participated in SemEval-2022 Task 6, iSarcasmEval - Intended Sarcasm Detection in English and Arabic, which aims at detecting intended sarcasm in various settings of natural language understanding. Our solution finetunes pre-trained language models, such as ERNIE-M and DeBERTa, under the multilingual settings to recognize the irony from Arabic and English texts. Our system ranked second out of 43, and ninth out of 32 in Task A: one-sentence detection in English and Arabic; fifth out of 22 in Task B: binary multi-label classification in English; first out of 16, and fifth out of 13 in Task C: sentence-pair detection in English and Arabic.