Video Compression Meets Video Generation: Latent Inter-Frame Pruning with Attention Recovery

Documenti analoghi

• Mitigating Intra- and Inter-modal Forgetting in Continual Learning of Unified Multimodal Models
  di: Wei, Xiwen, et al.
  Pubblicazione: (2025)
• Fuel Gauge: Estimating Chain-of-Thought Length Ahead of Time in Large Multimodal Models
  di: Yang, Yuedong, et al.
  Pubblicazione: (2026)
• Ada-VE: Training-Free Consistent Video Editing Using Adaptive Motion Prior
  di: Mahmud, Tanvir, et al.
  Pubblicazione: (2024)
• Synthesizing Artifact Dataset for Pixel-level Detection
  di: Menn, Dennis, et al.
  Pubblicazione: (2025)
• Latent Inter-Frame Pruning: A Training-Free Method Bridging Traditional Video Compression and Modern Diffusion Transformers for Efficient Generation
  di: Menn, Dennis, et al.
  Pubblicazione: (2026)