


Latest move
World Labs → AMD
Path
Current
AMDNEWJoined
Previous
World Labs– 2026Cofounder
Papers
- Perceptual Losses for Real-Time Style Transfer and Super-ResolutionarXiv 2016Trains feed-forward image transformation networks with losses computed from a pretrained network's features, for fast style transfer and super-resolution.
- CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual ReasoningarXiv 2016Builds a synthetic dataset of rendered 3D scenes and questions to test which visual reasoning skills a model has.
- Image Generation from Scene GraphsCVPR 2018 2018Generates images from scene graphs, structured descriptions of objects and their relations, rather than from sentences.
Interviews and talks
- 2026-09-04Fei Fei Li: The Race to Build World Models
For AI · The a16z Show · Podcast ↗ - 2026-09-01World Models and the Future of Spatial AI with
Justin Johnson · The TWIMLAI Podcast · Podcast ↗ - 2025-12-05What Comes After ChatGPT? The Mother of ImageNet Predicts
The Future · The a16z Show · Podcast ↗ - 2025-11-25After LLMs: Spatial Intelligence and World Models — Fei-Fei Li & Justin Johnson,
World Labs · Latent Space: The AIEngineer Podcast · Podcast ↗
SourcesLatest move: X