← Back to Model Beat
4Research·5h ago

QWERTY: Training-Free Motion Control via Query-Warped Video Diffusion Transformers

Researchers have introduced QWERTY, a method for controlling movement in video generation models without needing additional training or extensive prompt tuning. This approach uses query-warping to guide video diffusion transformers, offering users more precise command over the direction and speed of generated motion.

Covered by 1 source

  • AarXiv CS.AIKyobin Choo, Youngmin Kim, Hyunkyung Han, Geunrip Park, Chanyoung Kim, Sunyoung Jung, Seong Jae Hwang5h ago

Related stories

ResearchOn Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMsJun 29 · 13 sourcesResearchAnti-Causal Domain Generalization: Leveraging Unlabeled DataJul 1 · 2 sourcesResearchLearning Unmasking Policies for Diffusion Language ModelsJun 29 · 6 sourcesResearchRedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttentionJun 29