AI Paper Atlas

Research map / Video / Video generation

Reward Lightning: Fast Video Generation via Homologous Preference Distillation

Jiaxiang Cheng, bing ma, Xuhua Ren, Kai Yu, Peng Zhang, Tianxiang Zheng, Qinglin Lu

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
add few-step augmentation to LRM training data (ablation)LRM · VideoGen-RewardBench (LRM ablation)accuracy: 68.47 -> 72.24Improved
add real-synthetic pairs to LRM training data (ablation)LRM · VideoGen-RewardBench (LRM ablation)accuracy: 65.12 -> 72.24Improved
use time-step projection head instead of self-attention head (ablation)LRM · VideoGen-RewardBench (LRM ablation)accuracy: 64.38 -> 72.24Improved
use time-step projection head instead of trainable query head (ablation)LRM · VideoGen-RewardBench (LRM ablation)accuracy: 69.85 -> 72.24Improved
add reward regularization (rho=0.95) (ablation)LRM · VideoGen-RewardBench (LRM ablation)accuracy: 62.15 -> 72.24Improved
initialize discriminator with LRM instead of random (ablation)HPD · VBench-I2V (4-NFE)I2V Score: 88.26 -> 95.34Improved
use adaptive preference weight instead of lambda=0 (ablation)HPD · VBench-I2V (4-NFE)I2V Score: 90.87 -> 95.34Improved
share backbone between reward and distillation (homology) (ablation)HPD · Gradient conflict (4-NFE I2V)cosine similarity: -0.14 -> 0.41Improved
add LRM to rCMrCM · VBench-I2V (4-NFE)I2V Score: 93.45 -> 94.32Improved
add LRM to DMD2DMD2 · VBench-I2V (4-NFE)I2V Score: 91.92 -> 92.71Improved
add LRM to HPD (ablation)HPD · VBench-I2V (4-NFE)FVD: 167.14 -> 162.87Improved
use BTT loss instead of BT loss (ablation)LRM · VideoGen-RewardBenchaccuracy w/o Ties: 71.89 -> 72.24Improved

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