Comparar métodos

	Perceptron Multicamadas Multimodal	Transformer Multimodal
Área	Aprendizado profundo	Aprendizado profundo
Família	Machine learning	Machine learning
Ano de origem	2011 (multimodal extension); 1986 (MLP backpropagation)	2019–2021
Autor original	Ngiam et al. / Rumelhart, Hinton & Williams (MLP foundations)	Lu et al. (ViLBERT); Radford et al. (CLIP)
Tipo	Feedforward neural network with multi-stream fusion	Cross-modal attention-based deep learning model
Fonte seminal	Ngiam, J., Khosla, A., Kim, M., Nam, J., Lee, H., & Ng, A. Y. (2011). Multimodal deep learning. In Proceedings of the 28th International Conference on Machine Learning (ICML 2011), pp. 689–696.	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.
Outros nomes	MM-MLP, multimodal MLP, multi-input feedforward network, fusion multilayer perceptron	multimodal attention model, cross-modal transformer, vision-language transformer, multi-modal fusion transformer
Relacionados	5	5
Resumo	A Multimodal Multilayer Perceptron (MM-MLP) is a feedforward neural network that ingests features from two or more heterogeneous input modalities — such as structured tabular data, text embeddings, and image feature vectors — by encoding each stream separately and fusing them into a shared representation before passing it through fully connected layers to produce a classification or regression output.	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.