Research map / 3D vision / 3D generation
Edit-As-Act: Goal-Regressive Planning for Open-Vocabulary 3D Indoor Scene Editing
Seongrae Noh, SeungWon Seo, Gyeong-Moon Park, HyeongYeop Kang
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 |
|---|---|---|---|
| goal-regressive symbolic planning instead of single-pass LLM reasoning (GPT-5) | GPT-5 · E2A-Bench | IF/SC/PP: 49.6/52.3/73.3 -> 69.1/86.6/91.7 | Improved |
| Edit-As-Act instead of SceneWeaver (same GPT-5) | GPT-5 editing agent · E2A-Bench | IF/SC/PP: SC 78.3 -> 86.6; PP 82.1 -> 91.7; IF 68.7 -> 69.1 | Improved |
| add validator module (ablation) | Edit-As-Act · E2A-Bench | SC: 75.1 -> 86.6 | Improved |
| add source-aware regression (ablation) | Edit-As-Act · E2A-Bench | IF: 58.2 -> 69.1 | Improved |
| add EditLang symbolic action language instead of generic scene graph (ablation) | Edit-As-Act · E2A-Bench | SC: 73.6 -> 86.6 | Improved |
| backward goal regression instead of forward planning (ablation) | Edit-As-Act · E2A-Bench | IF: 61.2 -> 69.1 | Improved |
| symbolic planning instead of direct 3D coordinate prediction (ablation) | Edit-As-Act · E2A-Bench | IF: 52.8 -> 69.1 | Improved |
| Edit-As-Act versus ArtiScene-E on physical plausibility in the Bedroom/Office categories | ArtiScene-E · E2A-Bench Office | PP: 92.9 (ArtiScene-E) vs 81.9 (Edit-As-Act) | Mixed |
| Edit-As-Act cost versus SceneWeaver | Edit-As-Act · E2A-Bench | Avg model calls: 7.3 -> 5.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- AnyHome — AnyHome: Open-Vocabulary Large-Scale Indoor Scene Generation with First-Person View Exploration (ECCV 2024) · better on 5 of 5 shared results
- SceneWeaver — SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025) · better on 4 of 4 shared results
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
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.90)
- EditRoom: LLM-parameterized Graph Diffusion for Composable 3D Room Layout Editing (ICLR 2025 · similarity 0.90)
- SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025 · similarity 0.88)
- Hierarchically-Structured Open-Vocabulary Indoor Scene Synthesis with Pre-trained Large Language Model (AAAI 2025 · similarity 0.88)
- Learning Action and Reasoning-Centric Image Editing from Videos and Simulation (NeurIPS 2024 · similarity 0.87)
- 3D Space as a Scratchpad for Editable Text-to-Image Generation (CVPR 2026 · 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.
