Sarcasm Detection in Social Media: Techniques, Models, and Future Directions

DOI

10.2991/978-94-6463-823-3_93How to use a DOI?

Keywords

Sarcasm detection; Social media; Multimodal Transformer; Large Language Models; Contrastive Learning and Adversarial Training

Abstract

Sarcasm is extensively employed on social media to express complex or implicit emotional stances. Such expressions often create a discrepancy between literal meaning and actual sentiment, posing significant challenges to traditional sentiment analysis methods. This impacts their accuracy in practical applications like public opinion monitoring and brand management. This paper provides a systematic review of the mainstream approaches to sarcasm detection in social media developed over the past two years. These include multimodal Transformer-based architectures, techniques driven by Large Language Models (LLMs), and robustness enhancement methods based on contrastive learning and adversarial training. The strengths and limitations of these approaches are analyzed in terms of semantic understanding, modality fusion, and generalization ability. For instance, while multimodal Transformers excel in capturing contextual nuances, they may struggle with generalizing across diverse datasets. Similarly, LLM-driven methods show promise in understanding implicit sarcasm but require substantial computational resources. Future directions are also discussed, such as self-supervised learning for unlabeled data, advanced multimodal contextual modeling, and integrating emoji semantics to improve detection precision.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Haojie Song
PY  - 2025
DA  - 2025/08/31
TI  - Sarcasm Detection in Social Media: Techniques, Models, and Future Directions
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 945
EP  - 957
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_93
DO  - 10.2991/978-94-6463-823-3_93
ID  - Song2025
ER  -