Poster · Sydney
ActO: Extracting Action Representations from MLLM Embeddings for Video World Models
- When
- Tue 8 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 1
- Authors
- Runjia Li, Minghao Chen, Junyu Xie, Philip Torr, Andrea Vedaldi, Tomas Jakab
- Organisations
- Google DeepMind, University of Oxford, Meta
- QRS classification
- Computer vision