Scaling Video Generation for Reasoning: At What Cost?
A controlled Rubik's Cube benchmark reveals the compute cost of hidden-state reasoning and why lower generation loss does not guarantee better reasoning.
- video-generation
- reasoning
- scaling
Blog
Compress to Remember
Learning Compact Memory via On-Policy Distillation for Long Video Generation
Prediction-Aligned Context Compaction learns what a frozen video generator should remember. Explore the training method, compact-memory inference, and matched video comparisons.
- video generation
- memory
- distillation
Taming Outlier Tokens in Diffusion Transformers
Dual-Stage Registers reduce harmful outlier tokens in both vision encoders and diffusion transformers, improving generation quality across ImageNet and text-to-image settings.
- diffusion
- DiT
- registers
Leveraging Model Guidance to Extract Training Data from Personalized Diffusion Models
FineXtract uses the distribution shift between pretrained and personalized diffusion models to recover about 20% of fine-tuning data from real-world checkpoints.
- diffusion
- data-extraction
- privacy
Mist: Towards Improved Adversarial Examples for Diffusion Models
Mist improves the transferability and robustness of adversarial watermarks designed to protect artwork from unauthorized diffusion-model imitation.
- diffusion
- adversarial-examples
- artwork-protection