Saved in:
Bibliographic Details
Main Authors: Basnet, Saroj, Farabi, Shafkat, Ranasinghe, Tharindu, Kanoji, Diptesh, Zampieri, Marcos
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.11852
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917011026608128
author Basnet, Saroj
Farabi, Shafkat
Ranasinghe, Tharindu
Kanoji, Diptesh
Zampieri, Marcos
author_facet Basnet, Saroj
Farabi, Shafkat
Ranasinghe, Tharindu
Kanoji, Diptesh
Zampieri, Marcos
contents Recent advances in open-source vision-language models (VLMs) offer new opportunities for understanding complex and subjective multimodal phenomena such as sarcasm. In this work, we evaluate seven state-of-the-art VLMs - BLIP2, InstructBLIP, OpenFlamingo, LLaVA, PaliGemma, Gemma3, and Qwen-VL - on their ability to detect multimodal sarcasm using zero-, one-, and few-shot prompting. Furthermore, we evaluate the models' capabilities in generating explanations to sarcastic instances. We evaluate the capabilities of VLMs on three benchmark sarcasm datasets (Muse, MMSD2.0, and SarcNet). Our primary objectives are twofold: (1) to quantify each model's performance in detecting sarcastic image-caption pairs, and (2) to assess their ability to generate human-quality explanations that highlight the visual-textual incongruities driving sarcasm. Our results indicate that, while current models achieve moderate success in binary sarcasm detection, they are still not able to generate high-quality explanations without task-specific finetuning.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11852
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating Open-Source Vision-Language Models for Multimodal Sarcasm Detection
Basnet, Saroj
Farabi, Shafkat
Ranasinghe, Tharindu
Kanoji, Diptesh
Zampieri, Marcos
Machine Learning
Recent advances in open-source vision-language models (VLMs) offer new opportunities for understanding complex and subjective multimodal phenomena such as sarcasm. In this work, we evaluate seven state-of-the-art VLMs - BLIP2, InstructBLIP, OpenFlamingo, LLaVA, PaliGemma, Gemma3, and Qwen-VL - on their ability to detect multimodal sarcasm using zero-, one-, and few-shot prompting. Furthermore, we evaluate the models' capabilities in generating explanations to sarcastic instances. We evaluate the capabilities of VLMs on three benchmark sarcasm datasets (Muse, MMSD2.0, and SarcNet). Our primary objectives are twofold: (1) to quantify each model's performance in detecting sarcastic image-caption pairs, and (2) to assess their ability to generate human-quality explanations that highlight the visual-textual incongruities driving sarcasm. Our results indicate that, while current models achieve moderate success in binary sarcasm detection, they are still not able to generate high-quality explanations without task-specific finetuning.
title Evaluating Open-Source Vision-Language Models for Multimodal Sarcasm Detection
topic Machine Learning
url https://arxiv.org/abs/2510.11852