AI Paper Atlas

Research map / Video / Video generation

SSBP: Stage-Specialized Block Pruning for Video Diffusion Models

Mengmeng Ge, Takashi Isobe, Dong Zhou, Dong Li, Emad Barsoum

ECCV 2026 · Video › Video generation · approach: Efficient video diffusion models · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
partition denoising into stages and prune per stage instead of one global pruning (ablation)SSBP · VBench 30% accelerationtotal score: 78.07 -> 80.91Improved
add Inter-Stage Variance Selector (ablation)SSBP · VBench 30% accelerationtotal score: 80.91 -> 82.44Improved
add Intra-Stage Quality Selector (ablation)SSBP · VBench 30% accelerationtotal score: 82.44 -> 83.52Improved
use 4 stages instead of 1 (ablation)SSBP · VBench 30% accelerationtotal score: 80.16 -> 83.52Improved
use 4 stages instead of 2 (ablation)SSBP · VBench 20% accelerationtotal score: 82.70 -> 84.02Improved
use 3 candidates in IQS instead of 1 (ablation)SSBP · VBench 30% accelerationtotal score: 82.44 -> 83.52Improved
increase IQS candidates from 3 to 5 or 7 (ablation)SSBP · VBench 30% accelerationtotal score: 83.52 -> 83.63 (N=5), 83.60 (N=7)No clear effect
use SSBP instead of TinyFusionSSBP · VBench 30% accelerationtotal score: 80.46 -> 83.52Improved
use SSBP instead of L-OBSSSBP · VBench 20% accelerationtotal score: 79.62 -> 84.02Improved
prune 30% per-step cost versus unpruned DMD2 modelSSBP · VBenchtotal score: 84.21 -> 83.52Hurt
compare pruned model quality vs semantic score to DMD2SSBP · VBenchquality vs semantic score: quality 85.98 -> 85.02; semantic 77.09 -> 77.54Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy