AI Paper Atlas

Research map / Video / Video generation

Target-Aware Video Diffusion Models

Taeksoo Kim, Hanbyul Joo

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
add cross-attention loss on the [TGT] token (lambda_attn 0.0 -> 0.1) (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.647 -> 0.878Improved
fine-tune CogVideoX on the curated dataset only (no mask or loss)CogVideoX · Target-aware benchmarkContact Score: 0.560 -> 0.638Improved
apply loss to V2T cross-attention rather than T2V (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.740 -> 0.878Improved
apply loss to V2T only rather than both T2V and V2T (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.860 -> 0.878Improved
choose loss blocks by semantic alignment instead of random or equally spaced (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.819 (random), 0.816 (equally spaced) -> 0.878Improved
increase lambda_attn beyond 0.1 (0.25, 0.5, 1.0) (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.878 -> 0.890, 0.888, 0.888No clear effect
dilate or erode the target mask (ablation)Target-aware CogVideoX · Mask quality ablationContact Score: 0.896 -> 0.872 to 0.904No clear effect
replace mask with a circular mask (radius 15 or 30) (ablation)Target-aware CogVideoX · Mask shape ablationContact Score: 0.896 -> 0.838 (r15), 0.888 (r30)Mixed
use automatically generated masks (GPT-4o + Grounded-SAM) (ablation)Target-aware CogVideoX · Target-aware benchmark (50 images)Contact Score: 0.896 -> 0.864No clear effect
add captioner-generated descriptions to the general sentence in training prompts (ablation)Target-aware CogVideoX · Target-aware benchmarkContact Score: 0.705 -> 0.878Improved
selective loss on every 5th block in blocks 5-23Target-aware CogVideoX · VRAM usageVRAM usage reduction: 71% reduction vs all blocksImproved
add target-aware training vs attention modulation (Direct-a-video style)CogVideoX · Target-aware benchmarkContact Score: 0.560 -> 0.546Hurt

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