#artificial intelligence
Jan 2026
Turbo4DGen: Ultra-Fast Acceleration for 4D Generation
Turbo4DGen introduces a spatiotemporal cache mechanism that persistently reuses intermediate attention across denoising steps, combined with dynamically semantic-aware attention pruning and an adaptive SCM chain bypass scheduler, to drastically reduce redundant SCM attention computation.
Yuanbin Man, Ying Huang, Zhile Ren et al.
· arXiv.org · 0 citations