AI Paper Atlas

Research map / Video / Video generation

TTOM: Test-Time Optimization and Memorization for Compositional Video Generation

Leigang Qu, Ziyang Wang, Na Zheng, Wenjie Wang, Liqiang Nie, Tat-Seng Chua

ICLR 2026 · Video › Video generation · approach: Compositional text-to-video diffusion · 5 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
add test-time optimization with layout loss (ablation)Wan2.1-14B · T2V-CompBench MotionMotion score: 0.2696 -> 0.4321Improved
add parametric memory on top of TTO (ablation)TTOM · T2V-CompBench MotionMotion score: 0.4321 -> 0.4922Improved
add memory initialization (top-5) without TTO (ablation)Wan2.1-14B · T2V-CompBench MotionMotion score: 0.2696 -> 0.4754Improved
add continual TTO after memory init (top-5) (ablation)TTOM · T2V-CompBench MotionMotion score: 0.4754 -> 0.4846; latency 427 -> 627Improved
raise top-k memory entries fused for init from 5 to 10 (ablation)TTOM · T2V-CompBench MotionMotion score: 0.4754 -> 0.4437Hurt
use JSD loss instead of CE loss (ablation)TTOM · T2V-CompBench MotionMotion score: 0.2912 -> 0.4321Improved
use JSD loss instead of CoM loss (ablation)TTOM · T2V-CompBench NumeracyNumeracy score: 0.4697 -> 0.5881Improved
increase pseudo-training prompts for memory from 50 to 200 (ablation)TTOM · T2V-CompBench MotionMotion score: 0.3519 -> 0.4922Improved
increase optimized denoising timesteps from 1 to 5 (ablation)TTOM · T2V-CompBench MotionMotion score: 0.2730 -> 0.4321Improved
increase optimized timesteps from 5 to 7 (ablation)TTOM · T2V-CompBench MotionMotion score: 0.4321 -> 0.4296No clear effect
increase iterations per timestep from 8 to 16 (ablation)TTOM · T2V-CompBench MotionMotion score: 0.4321 -> 0.3547Hurt
apply TTOM to CogVideoX-5BCogVideoX-5B · T2V-CompBenchAvg score: 0.4189 -> 0.5632Improved

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