When did you become so smart, oh wise one?! Sarcasm Explanation in Multi-modal Multi-party Dialogues

TL;DR

This paper introduces a new task and dataset for explaining sarcasm in multi-modal, multi-party dialogues, emphasizing the importance of interpretability for conversational AI.

Contribution

It proposes the Sarcasm Explanation in Dialogue (SED) task, curates the WITS dataset, and develops the MAF multimodal fusion module that outperforms existing baselines.

Findings

MAF achieves superior performance on sarcasm explanation metrics.

WITS dataset enables effective training and evaluation of sarcasm explanation models.

Multimodal context-aware attention improves understanding of sarcastic dialogues.

Abstract

Indirect speech such as sarcasm achieves a constellation of discourse goals in human communication. While the indirectness of figurative language warrants speakers to achieve certain pragmatic goals, it is challenging for AI agents to comprehend such idiosyncrasies of human communication. Though sarcasm identification has been a well-explored topic in dialogue analysis, for conversational systems to truly grasp a conversation's innate meaning and generate appropriate responses, simply detecting sarcasm is not enough; it is vital to explain its underlying sarcastic connotation to capture its true essence. In this work, we study the discourse structure of sarcastic conversations and propose a novel task - Sarcasm Explanation in Dialogue (SED). Set in a multimodal and code-mixed setting, the task aims to generate natural language explanations of satirical conversations. To this end, we curate WITS, a new dataset to support our task. We propose MAF (Modality Aware Fusion), a multimodal context-aware attention and global information fusion module to capture multimodality and use it to benchmark WITS. The proposed attention module surpasses the traditional multimodal fusion baselines and reports the best performance on almost all metrics. Lastly, we carry out detailed analyses both quantitatively and qualitatively.