Efficient Visual Generation (EVG) brings together researchers and practitioners working on efficient generative models for visual and multimodal content. The series focuses on methods that make high-quality generation more practical across constrained devices, interactive applications, production systems, and large-scale deployment.
Across editions, EVG highlights architectural innovation, data-efficient learning, resource-aware inference, and deployment-ready optimization strategies for image, video, multimodal, and world-modeling systems.