Abstract
This paper provides a comparison of different deep learning methods for identifying misogynous memes for SemEval-2022 Task 5: Multimedia Automatic Misogyny Identification. In this task, we experiment with architectures in the identification of misogynous content in memes by making use of text and image-based information. The different deep learning methods compared in this paper are: (i) unimodal image or text models (ii) fusion of unimodal models (iii) multimodal transformers models and (iv) transformers further pretrained on a multimodal task. From our experiments, we found pretrained multimodal transformer architectures to strongly outperform the models involving the fusion of representation from both the modalities.- Anthology ID:
- 2022.semeval-1.92
- Volume:
- Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)
- Month:
- July
- Year:
- 2022
- Address:
- Seattle, United States
- Venue:
- SemEval
- SIGs:
- SIGLEX | SIGSEM
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 673–678
- Language:
- URL:
- https://aclanthology.org/2022.semeval-1.92
- DOI:
- 10.18653/v1/2022.semeval-1.92
- Cite (ACL):
- Tathagata Raha, Sagar Joshi, and Vasudeva Varma. 2022. IIITH at SemEval-2022 Task 5: A comparative study of deep learning models for identifying misogynous memes. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), pages 673–678, Seattle, United States. Association for Computational Linguistics.
- Cite (Informal):
- IIITH at SemEval-2022 Task 5: A comparative study of deep learning models for identifying misogynous memes (Raha et al., SemEval 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.semeval-1.92.pdf
- Data
- COCO, Hateful Memes, Visual Genome