Machine learningDeep learning / NLP / CV

Multimodalni Transformer

Multimodalni Transformer proširuje standardnu Transformer arhitekturu za obradu i zajedničko rezonovanje nad dve ili više ulaznih modaliteta — najčešće teksta i slika, ali takođe i zvuka, videa ili strukturiranih podataka. Slojevi unakrsne pažnje (cross-modal attention) omogućavaju informacijama iz jednog modaliteta da utiču na reprezentacije u drugom, omogućavajući zadatke kao što su vizuelno odgovaranje na pitanja, opisivanje slika i multimodalna analiza sentimenta.

Otvorite u MethodMindUskoroVideoUskoroDownload slides

Pročitajte celu metodu

Samo za članove

Prijavite se besplatnim nalogom da biste pročitali ovaj odeljak.

Prijavite se

Method map

The neighbourhood of related methods — select a node to explore.

+15 more

Izvori

  1. 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. link
  2. Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., ... & Sutskever, I. (2021). Learning Transferable Visual Models From Natural Language Supervision. Proceedings of the 38th International Conference on Machine Learning (ICML), PMLR 139. link

Kako citirati ovu stranicu

ScholarGate. (2026, June 3). Multimodal Transformer (Cross-Modal Attention-Based Architecture). ScholarGate. https://scholargate.app/sr/deep-learning/multimodal-transformer

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

Compare side by side

Citirana u

ScholarGateMultimodal Transformer (Multimodal Transformer (Cross-Modal Attention-Based Architecture)). Preuzeto 2026-06-15 sa https://scholargate.app/sr/deep-learning/multimodal-transformer · Skup podataka: https://doi.org/10.5281/zenodo.20539026