AI Paper Atlas

Research map / Video / Video-language understanding

Stitch-a-Demo: Creating Video Demonstrations from Multistep Descriptions

Chi Hsuan Wu, Kumar Ashutosh, Kristen Grauman

CVPR 2026 · Video › Video-language understanding · approach: Video-language pretraining

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 LLM-created weakly supervised mixed procedures to training (ablation)Stitch-a-Demo · SaD-VDR@5: 0.14 -> 0.56 (vs no augmentation)Improved
use LLM-created procedures instead of temporally-sampled combinations (ablation)Stitch-a-Demo · SaD-VDmedian rank: 10 -> 3.5Improved
add visual continuity negatives to correctness negatives (ablation)Stitch-a-Demo · SaD-VDR@5: 0.41 -> 0.45Improved
add object state negatives to correctness + continuity negatives (ablation)Stitch-a-Demo · SaD-VDR@5: 0.45 -> 0.52 (R@5); MedR 9 -> 5Mixed
use only visual continuity negatives (ablation)Stitch-a-Demo · SaD-VDR@5: 0.03 (MedR 171)Hurt
add correctness negatives to continuity + object state negatives (ablation)Stitch-a-Demo · SaD-VDR@5: 0.21 -> 0.52Improved
add visual continuity negatives to correctness + object state negatives (ablation)Stitch-a-Demo · SaD-VDR@5: 0.54 -> 0.52 (R@5); 0.15 -> 0.22 (R@1)Mixed
replace heuristic scoring (InternVideo similarity) with learned procedure evaluatorStitch-a-Demo · Cooking SaD-MCR@50: 0.55 -> 0.84Improved
apply Stitch-a-Demo to woodworking and gardeningStitch-a-Demo · Woodworking/Gardening SaD-MCR@50: 0.36 vs CoVR 0.34 (woodworking)Mixed
stitch clips from multiple videos instead of retrieving original single video demonstrationStitch-a-Demo · Cooking human preferencepreference ratio: 83:17Improved
use retrieval instead of image generation (ShowHowTo)Stitch-a-Demo · Cooking human preferencewin rate, total: 0.98Improved
compare VBench subject consistency at step transitions vs InternVideoStitch-a-Demo · object-state consistencyVBench subject consistency: 0.62 -> 0.67Improved

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