Metraipolis › archive › dc_c7ddae998092
Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification — arXiv:2609.26907
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.26907
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:05:27.285Z
- SHA-256 of the archived text
a3af40edb2de1becf7d6994fc7069c325731db39fe848f9e8fd47c11a5c20a10- Stable URL of this page
https://metraipolis.com/documento/dc_c7ddae998092
How to cite
Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification — arXiv:2609.26907 — arXiv (en), https://arxiv.org/abs/2609.26907 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c7ddae998092 (Metraipolis archive dc_c7ddae998092, sha256 a3af40edb2de1becf7d6994fc7069c325731db39fe848f9e8fd47c11a5c20a10)
BibTeX
@misc{dcc7ddae998092,
title = {Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification — arXiv:2609.26907},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.26907},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c7ddae998092; sha256 a3af40edb2de1becf7d6994fc7069c325731db39fe848f9e8fd47c11a5c20a10},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.26907. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Memes often derive their harmful, hateful, or sarcastic meaning from small but decisive visual, textual, or cross-modal cues. Existing multimodal classifiers can miss such evidence when relying mainly on global image-text representations. We introduce MemeCF, a cue-focused benchmark of 9,895 memes across harm, hate, and sarcasm, with annotations identifying the modality and rationale of the pivotal evidence. We also propose MemePIVOT, a local-global architecture for meme classification. MemePIVOT uses frozen CLIP features, unbalanced optimal transport to align words with image patches while allowing irrelevant evidence to remain unmatched, and an evidential fusion head to combine local grounding with global meme context under uncertainty. Experiments on HarMeme, PrideMM, and MemeCF show consistent gains over strong text-only, image-only, multimodal, and vision-language baselines. Cross-dataset and ablation results further show that explicit pivotal-evidence modeling improves robustness and contributes meaningfully beyond global multimodal representations. Our code and dataset are publicly available at this https URL Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.CL). Categorie arXiv: Multimedia (cs.MM) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.26907 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T09:11:49.671Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).