Hadar Averbuch-Elor

LLM interrogations reveal diffusion model's internal tokens capture scene semantics surprisingly early, even without explicit prompts, and correlate with generation quality.

Recovering lost visual details in long-horizon video generation is now possible by compositing non-contiguous spatio-temporal memory, even with a frozen generator.

Evaluating various models across diverse categories reveals that current models can only maintain short-term consistency that degrades significantly once objects leave the field of view, highlighting the gap between current model capabilities and robust visual memory ("seeing is not remembering"), providing guidance for future development of 4D foundation models.