AI Paper Atlas

Research map / Video / Video generation

Frame Guidance: Training-Free Guidance for Frame-Level Control in Video Diffusion Model

Sangwon Jang, Taekyung Ki, Jaehyeong Jo, Jaehong Yoon, Soo Ye Kim, Zhe Lin, Sung Ju Hwang

ICLR 2026 · Video › Video generation · approach: Efficient video diffusion models

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
use video latent optimization (deterministic early, stochastic later) instead of time-travel only (ablation)Frame Guidance · keyframe-guided generationFVD: 778.4 -> 577.1 FVD; 57.37 -> 55.60 FIDImproved
add stochastic later-stage updates to deterministic-only updates (full VLO) (ablation)Frame Guidance · keyframe-guided generationFVD: 637.3 -> 577.1 FVD; 56.61 -> 55.60 FIDImproved
decode only a 3-latent slice for the guidance loss (latent slicing) (ablation)Frame Guidance on CausalVAE · GPU memorymemory: up to 15x reductionImproved
add 2x spatial down-sampling of latents to latent slicing (ablation)Frame Guidance · GPU memorymemory: up to 60x reductionImproved
backpropagate guidance loss through the denoising network instead of shortcut update (ablation)Frame Guidance · keyframe-guided generationtemporal coherence (qualitative)Improved
use a 2-latent slice instead of 3-latent slice for reconstruction (ablation)latent slicing · video reconstructionreconstruction quality (qualitative)Hurt
vary t_E (end of deterministic stage) (ablation)VLO · DAVIS (20 videos) keyframe generationFID/FVD: best at t_E=6No clear effect
apply Frame Guidance to pre-trained I2V models for keyframe guidanceCogX and Wan · DAVIS and Pexels keyframe generationFID/FVDImproved
apply Frame Guidance on top of CogX-InterpCogX-Interp · keyframe-guided generationFID/FVDImproved
apply Frame Guidance to Wan I2VWan-14B I2V · keyframe-guided generationFID/FVDImproved
guide Sobel edge maps instead of RGB keyframesFrame Guidance · edge-map guidancecontrol quality (qualitative)Hurt

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