AI Paper Atlas

Research map / Video / Video generation

LatSearch: Latent Reward-Guided Search for Faster Inference-Time Scaling in Video Diffusion

Zengqun Zhao, Ziquan Liu, Yu Cao, Shaogang Gong, Zhensong Zhang, Song Jifei, Jiankang Deng, ioannis Patras

ECCV 2026 · Video › Video generation · approach: Efficient video diffusion models · 6 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
apply LatSearch latent reward-guided searchWan2.1-1.3B · VBench-2.0 averagescore: 51.90 -> 53.84Improved
LatSearch vs EvoSearchLatSearch · VBench-2.0score and time: 55.25 vs 55.01 average; 164.41 s vs 783.76 sMixed
initial-noise optimisation (FreeInit, FreqPrior)Wan2.1-1.3B · VBench-2.0 averagescore: 51.90 -> 49.82 / 50.37Hurt
increase search budget N from 4 to 8 (ablation)LatSearch · VBench-2.0 averagescore: 52.81 -> 54.13Improved
add RGRP resampling instead of beam-style selection (ablation)LatSearch · VBench-2.0 averagescore: 52.35 -> 53.84Improved
add preference loss to latent reward model (ablation)LatSearch · VBench-2.0 averagescore: 53.42 -> 53.84Improved
cosine similarity credit assignment instead of uniform/exponential/L2 (ablation)LatSearch · VBench-2.0 averagescore: 52.31 -> 53.84Improved
use scoring schedule {10,15,20} instead of five points (ablation)LatSearch · VBench-2.0 averagescore: 52.34 -> 53.84Improved
use scoring schedule {10,15,20} instead of {10,15} (ablation)LatSearch · VBench-2.0 averagescore: 52.17 -> 53.84Improved
vary temperature tau 0.5, 1.0, 2.0 (ablation)LatSearch · VBench-2.0 averagescore: 53.56 / 53.84 / 53.33No clear effect
apply LatSearch to CogVideoX-2BCogVideoX-2B · VBench-2.0 averagescore: 45.13 -> 48.14Improved
apply LatSearch to Wan2.1-14BWan2.1-14B · VBench-2.0 averagescore: 52.58 -> 53.61Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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