Comparer des méthodes

	Modèle thématique LDA multimodal	Classification multimodale basée sur BERT
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2003	2019
Auteur d'origine	Blei, D. M. & Jordan, M. I.	Kiela, D. et al.; Lu, J. et al.
Type	Probabilistic generative topic model (multimodal)	Multimodal transformer classifier
Source fondatrice	Blei, D. M. & Jordan, M. I. (2003). Modeling annotated data. Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 127–134.	Kiela, D., Bhooshan, S., Firooz, H., Perez, E., & Testuggine, D. (2019). Supervised multimodal bitransformers for classifying images and text. arXiv preprint arXiv:1909.02950.
Alias	Multimodal LDA, mm-LDA, multimodal topic model, cross-modal LDA	MMBT, multimodal transformer classification, BERT multimodal fusion, vision-language BERT classifier
Apparentées	6	2
Résumé	Multimodal LDA extends Latent Dirichlet Allocation to jointly model multiple data modalities — most often text and images — within a single probabilistic topic framework. Each document or data instance is represented as a mixture of latent topics shared across modalities, enabling the model to discover coherent themes that align visual and linguistic content simultaneously.	Multimodal BERT-based classification extends the BERT transformer architecture to jointly encode and classify data from multiple modalities — most commonly text paired with images — by fusing their representations before a final classification head. Introduced prominently around 2019 through models such as MMBT and ViLBERT, it has become a standard approach for tasks where neither text nor image alone carries sufficient information for accurate labeling.