Σύγκριση μεθόδων

	Μοντέλο Θεμάτων Πολυτροπικού LDA	Πολυτροπικός Μετασχηματιστής
Πεδίο	Βαθιά Μάθηση	Βαθιά Μάθηση
Οικογένεια	Machine learning	Machine learning
Έτος προέλευσης	2003	2019–2021
Δημιουργός	Blei, D. M. & Jordan, M. I.	Lu et al. (ViLBERT); Radford et al. (CLIP)
Τύπος	Probabilistic generative topic model (multimodal)	Cross-modal attention-based deep learning model
Θεμελιώδης πηγή	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.	Lu, J., Batra, D., Parikh, D., & Lee, S. (2019). ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks. Advances in Neural Information Processing Systems (NeurIPS), 32.
Εναλλακτικές ονομασίες	Multimodal LDA, mm-LDA, multimodal topic model, cross-modal LDA	multimodal attention model, cross-modal transformer, vision-language transformer, multi-modal fusion transformer
Συναφείς	6	5
Σύνοψη	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.	A Multimodal Transformer extends the standard Transformer architecture to process and jointly reason over two or more input modalities — most commonly text and images, but also audio, video, or structured data. Cross-modal attention layers allow information from one modality to inform representations in another, enabling tasks such as visual question answering, image captioning, and multimodal sentiment analysis.