Research map / Video / Video generation
EcoVideo: Entropy-Orchestrated Video Generation Paradigm in Cloud-Edge Dynamics
Jiayu Chen, Hengyi Zhang, Maoliang Li, Minyu Li, Zihao Zheng, Xuanzhe Liu, Guojie Luo, Xiang Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add entropy-based keyframe selection (vs uniform keyframes) (ablation) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.832 -> 0.845 | Improved |
| add entropy-based keyframe selection (vs uniform keyframes) (ablation) | EcoVideo · VBench, Wan2.1-14B | speedup: 1.90x -> 1.84x | Hurt |
| add EcoVFI edge reconstruction (vs keyframes only) (ablation) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.835 -> 0.845 | Improved |
| add EcoVFI edge reconstruction (vs keyframes only) (ablation) | EcoVideo · VBench, Wan2.1-14B | speedup: 2.02x -> 1.84x | Hurt |
| replace original EDEN interpolation with EcoVFI (ablation) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.835 -> 0.845 | Improved |
| lengthen warm-up from 10% to 30% (ablation) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.847; speedup 1.49x | Mixed |
| shorten warm-up to 2% (ablation) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.834; speedup 2.13x | Mixed |
| inter-frame decoupling instead of inter-step (EcoVideo vs EC-Diff) | EcoVideo · VBench, Wan2.1-14B | VBench Overall: 0.683 -> 0.846 | Improved |
| inter-frame decoupling instead of inter-step (EcoVideo vs EC-Diff) | EcoVideo · VBench, CogVideoX-5B | speedup: 0.98x -> 2.03x | Improved |
| inter-frame decoupling vs cloud-only | EcoVideo · VBench, Wan2.2-A14B | VBench Overall: 0.842 -> 0.830 | Hurt |
| inter-frame decoupling vs cloud-only | EcoVideo · VBench, CogVideoX-5B | VBench Overall: 0.819 -> 0.775 | Hurt |
| dynamic (K,D) adaptation under cloud contention vs fixed step splitting (ablation) | EcoVideo · Resource ablation | speedup: 1.04x, 1.68x -> 2.91x | Improved |
Closest papers by meaning
- AdaState: Self-Evolving Anchors for Streaming Video Generation (NeurIPS 2026 · similarity 0.85)
- InfinityStar: Unified Spacetime AutoRegressive Modeling for Visual Generation (NeurIPS 2025 · similarity 0.85)
- DynamicScaler: Seamless and Scalable Video Generation for Panoramic Scenes (CVPR 2025 · similarity 0.84)
- Foresight: Adaptive Layer Reuse for Accelerated and High-Quality Text-to-Video Generation (NeurIPS 2025 · similarity 0.84)
- Training-Free Efficient Video Generation via Dynamic Token Carving (NeurIPS 2025 · similarity 0.84)
- Equilibrium Forcing: Adaptive Video Generation Without Noise Conditioning (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
