DualParal parallelizes both temporal frames and DiT layers using block-wise denoising, KV caching, and coordinated noise initialization, generating 1,025-frame videos up to 6.54x faster than baselines.
Cogvideo: Large-scale pretraining for text-to-video generation via transformers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Minute-Long Videos with Dual Parallelisms
DualParal parallelizes both temporal frames and DiT layers using block-wise denoising, KV caching, and coordinated noise initialization, generating 1,025-frame videos up to 6.54x faster than baselines.