MemeLens: Multilingual Multitask VLMs for Memes
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909008979296256 |
|---|---|
| author | Shahroor, Ali Ezzat Kmainasi, Mohamed Bayan Hasnat, Abul Dimitrov, Dimitar Martino, Giovanni Da San Nakov, Preslav Alam, Firoj |
| author_facet | Shahroor, Ali Ezzat Kmainasi, Mohamed Bayan Hasnat, Abul Dimitrov, Dimitar Martino, Giovanni Da San Nakov, Preslav Alam, Firoj |
| contents | Memes are a dominant medium for online communication and manipulation because meaning emerges from interactions between embedded text, imagery, and cultural context. Existing meme research is distributed across tasks (hate, misogyny, propaganda, sentiment, humour) and languages, which limits cross-domain generalization. To address this gap we propose MemeLens, a unified multilingual and multitask explanation-enhanced Vision Language Model (VLM) for meme understanding. We consolidate $38$ public meme datasets, filter and map dataset-specific labels into a shared taxonomy of $20$ tasks spanning harm, targets, figurative/pragmatic intent, and affect. We present a comprehensive empirical analysis across modeling paradigms, task categories, and datasets. Our findings suggest that robust meme understanding requires multimodal training, exhibits substantial variation across semantic categories, and remains sensitive to over-specialization when models are fine-tuned on individual datasets rather than trained in a unified setting. We make the experimental resources (https://github.com/MohamedBayan/MemeLens), model (https://huggingface.co/QCRI/MemeLens-VLM) and datasets (https://huggingface.co/datasets/QCRI/MemeLens) publicly available to the community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12539 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MemeLens: Multilingual Multitask VLMs for Memes Shahroor, Ali Ezzat Kmainasi, Mohamed Bayan Hasnat, Abul Dimitrov, Dimitar Martino, Giovanni Da San Nakov, Preslav Alam, Firoj Artificial Intelligence Computation and Language 68T50 I.2.7 Memes are a dominant medium for online communication and manipulation because meaning emerges from interactions between embedded text, imagery, and cultural context. Existing meme research is distributed across tasks (hate, misogyny, propaganda, sentiment, humour) and languages, which limits cross-domain generalization. To address this gap we propose MemeLens, a unified multilingual and multitask explanation-enhanced Vision Language Model (VLM) for meme understanding. We consolidate $38$ public meme datasets, filter and map dataset-specific labels into a shared taxonomy of $20$ tasks spanning harm, targets, figurative/pragmatic intent, and affect. We present a comprehensive empirical analysis across modeling paradigms, task categories, and datasets. Our findings suggest that robust meme understanding requires multimodal training, exhibits substantial variation across semantic categories, and remains sensitive to over-specialization when models are fine-tuned on individual datasets rather than trained in a unified setting. We make the experimental resources (https://github.com/MohamedBayan/MemeLens), model (https://huggingface.co/QCRI/MemeLens-VLM) and datasets (https://huggingface.co/datasets/QCRI/MemeLens) publicly available to the community. |
| title | MemeLens: Multilingual Multitask VLMs for Memes |
| topic | Artificial Intelligence Computation and Language 68T50 I.2.7 |
| url | https://arxiv.org/abs/2601.12539 |