-
Video Generation Models are General-Purpose Vision Learners
GenCeption
-
Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
Self-Flow, FLUX 3
-
AIDE²: The First Evidence of Recursive Self-Improvement
RSI ish
-
Training Agents Inside of Scalable World Models
Dreamer v4
-
What’s in the Image? A Deep-Dive into the Vision of Vision Language Models
VLM QA mech interp