Мультимодальное действие

AbstractNext generation virtual assistants are envisioned to handle multimodal inputs (e.g., vision, memories of previous interactions, and the user’s utterances), and perform multimodal actions (, displaying a route while generating …...