[S5E2] Video Models Are Zero-Shot Learners and Reasoners | Thaddäus Wiedemer | Google Deepmind
Автор: The AI Talks
Загружено: 2025-11-03
Просмотров: 761
Описание:
Can video models, like LLMs, evolve into general-purpose foundation models? Wiedemer presents work showing how Veo 3 can zero-shot segment, edit, simulate, and reason — hinting at the emergence of unified visual intelligence.
🎙️Speaker: Thaddäus Wiedemer (https://thaddaeuswiedemer.github.io/), PhD student at the International Max Planck Research School for Intelligent Systems, currently interning at Google DeepMind Toronto.
🔗 Subscribe now: http://theaitalks.org
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке: