Research map / Video / Video generation
VEGGIE: Instructional Editing and Reasoning Video Concepts with Grounded Generation
Shoubin Yu, Difan Liu, Ziqiao Ma, Yicong Hong, Yang Zhou, Hao Tan, Joyce Chai, Mohit Bansal
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add multi-task training with grounding data to removal (ablation) | VEGGIE · VPLM removal | FVD: 1098.52 -> 987.80 | Improved |
| add multi-task training with removal data to grounding (ablation) | VEGGIE · VPLM grounding | SSIM: 52.34 -> 55.21 | Improved |
| add excessive grounding data to the mixture (ablation) | VEGGIE · video editing | editing artifacts | Hurt |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench Concept Removal | MLLM-Judge: 2.78 -> 5.07 | Improved |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench Object Grounding | Jaccard index J: 13.89 -> 37.74 | Improved |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench Reasoning Segmentation | Jaccard index J: 16.89 -> 22.53 | Improved |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench Visual Feature Changing | MLLM-Judge: 7.53 -> 7.33 | Hurt |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench Concept Addition | MLLM-Judge: 5.69 -> 7.44 | Improved |
| use VEGGIE instead of InsV2V | instructional editing · VEG-Bench all skills | Average ranking: 2.00 -> 1.78 | Improved |
| use two-stage curriculum (image alignment first) instead of training directly on all data (ablation) | VEGGIE · training | training stability | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Flatten — FLATTEN: optical FLow-guided ATTENtion for consistent text-to-video editing (ICLR 2024) · better on 1 of 1 shared result
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
- Señorita-2M: A High-Quality Instruction-based Dataset for General Video Editing by Video Specialists (NeurIPS 2025 · similarity 0.88)
- VEU-Bench: Towards Comprehensive Understanding of Video Editing (CVPR 2025 · similarity 0.87)
- VicEdit: Learning to Edit Videos from Visual In-Context Examples (NeurIPS 2026 · similarity 0.87)
- Learning Action and Reasoning-Centric Image Editing from Videos and Simulation (NeurIPS 2024 · similarity 0.87)
- Text-Vision Co-Instructed Image Editing (NeurIPS 2026 · similarity 0.87)
- Video-STaR: Self-Training Enables Video Instruction Tuning with Any Supervision (ICLR 2025 · similarity 0.87)
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.
