GDCNet: Generative Discrepancy Comparison Network for Multimodal Sarcasm Detection

TL;DR

GDCNet is a novel framework for multimodal sarcasm detection that leverages factually grounded image captions from Multimodal LLMs to measure semantic and sentiment discrepancies, improving accuracy and robustness.

Contribution

The paper introduces GDCNet, which uses grounded image captions as semantic anchors to better detect sarcasm across image-text pairs, addressing limitations of previous methods.

Findings

Achieves state-of-the-art accuracy on MSD benchmarks.

Effectively captures semantic and sentiment discrepancies.

Demonstrates robustness in diverse multimodal scenarios.

Abstract

Multimodal sarcasm detection (MSD) aims to identify sarcasm within image-text pairs by modeling semantic incongruities across modalities. Existing methods often exploit cross-modal embedding misalignment to detect inconsistency but struggle when visual and textual content are loosely related or semantically indirect. While recent approaches leverage large language models (LLMs) to generate sarcastic cues, the inherent diversity and subjectivity of these generations often introduce noise. To address these limitations, we propose the Generative Discrepancy Comparison Network (GDCNet). This framework captures cross-modal conflicts by utilizing descriptive, factually grounded image captions generated by Multimodal LLMs (MLLMs) as stable semantic anchors. Specifically, GDCNet computes semantic and sentiment discrepancies between the generated objective description and the original text, alongside measuring visual-textual fidelity. These discrepancy features are then fused with visual and textual representations via a gated module to adaptively balance modality contributions. Extensive experiments on MSD benchmarks demonstrate GDCNet's superior accuracy and robustness, establishing a new state-of-the-art on the MMSD2.0 benchmark.