The Seriality Gap in Video Diffusion Models
When one ball strikes another, then another, video models should predict the consequences of each bounce. In controlled experiments on multi-ball hard-sphere dynamics, we find that the performance of standard bidirectional video diffusion degrades as the causal chain lengthens, even when provided more denoising steps. In a length-matched single-ball control, where ball-ball interactions are absent, the degradation largely disappears, isolating dependent-event structure rather than video length as the cause. Across intervention studies, methods that increase effective serial computation improve
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 66%stabilityai/stable-video-diffusion-img2vid-xt →
“Fuzzy title match (0.78): “The Seriality Gap in Video Diffusion Models” ≈ “stabilityai/stable-video-diffusion-img2vid-xt””
- FuzzySimilar title/name (fuzzy) · 59%CompVis/stable-diffusion-v1-4 →
“Fuzzy title match (0.73): “The Seriality Gap in Video Diffusion Models” ≈ “CompVis/stable-diffusion-v1-4””
- LinkedLinked via arxiv author · 85%Jorge Diaz Chao →
“The Seriality Gap in Video Diffusion Models”
- LinkedLinked via arxiv author · 85%Konpat Preechakul →
“The Seriality Gap in Video Diffusion Models”
- LinkedLinked via arxiv author · 85%Yuxi Liu →
“The Seriality Gap in Video Diffusion Models”
- LinkedLinked via arxiv author · 85%Yutong Bai →
“The Seriality Gap in Video Diffusion Models”
- FuzzyOverlapping authors or contributors · 62%modular/modular →
“Shared author/contributor keys: liu”
- FuzzySimilar title/name (fuzzy) · 59%Developer-Y/cs-video-courses →
“Fuzzy title match (0.73): “The Seriality Gap in Video Diffusion Models” ≈ “Developer-Y/cs-video-courses””