Research map / 3D vision / 3D generation
LabBuilder: Protocol-Grounded 3D Layout Generation for Interactable and Safe Laboratory
Jianbao Cao, zhangrui zhao, Bohan Feng, Zixuan Hu, Rui Li, Haiyuan Wan, Chenxi Li, jingyuan Li, Wenzhe Cai, LEI BAI, Wanli Ouyang, LINGYU DUAN and 6 more
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 structured protocol guidance (ablation) | LabBuilder · LabTouchstone, 30 experiments | asset availability: 0.300 -> 0.833 | Improved |
| add structured protocol guidance (ablation) | LabBuilder · LabTouchstone, 30 experiments | boundary violations (OB): 2.80 -> 0.07 | Improved |
| add semantic and domain asset annotations (ablation) | LabBuilder · LabTouchstone, 30 experiments | flammable isolation: 0.452 -> 0.725 | Improved |
| add semantic and domain asset annotations (ablation) | LabBuilder · LabTouchstone, 30 experiments | LLM completeness: 5.54 -> 9.10 | Improved |
| add geometric and chemical optimization loop (ablation) | LabBuilder · LabTouchstone, 30 experiments | OB/CN violations: 2.73/1.43 -> 0.07/0.17 | Improved |
| add geometric and chemical optimization loop (ablation) | LabBuilder · LabTouchstone, 30 experiments | incompatibility separation: 0.367 -> 0.716 | Improved |
| add navigation-aware refinement (ablation) | LabBuilder · LabTouchstone, 30 experiments | protocol-level navigation: 0.533 -> 0.867 | Improved |
| add navigation-aware refinement (ablation) | LabBuilder · LabTouchstone, 30 experiments | chemistry and geometry metrics: OB/CN 0.07/0.17 vs. 0.07/0.17 | No clear effect |
| add navigation-aware refinement to scenes | NavDP · Point-goal navigation in generated labs | success rate: 0.439 -> 0.517 | Improved |
| add navigation-aware refinement to scenes | iPlanner · Point-goal navigation in generated labs | success rate: 0.350 -> 0.425 | Improved |
| give Holodeck the same structured protocol info (Holodeck+) | Holodeck · LabTouchstone, 30 experiments | flammable isolation / boundary violations: Flam. 0.239 -> 0.723; OB 10.8 -> 15.2 | Mixed |
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.
- Holodeck — Holodeck: Language Guided Generation of 3D Embodied AI Environments (CVPR 2024) · better on 13 of 13 shared results
- SceneWeaver — SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025) · better on 3 of 3 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
- Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning (NeurIPS 2025 · similarity 0.83)
- LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents (NeurIPS 2025 · similarity 0.83)
- Layout-your-3D: Controllable and Precise 3D Generation with 2D Blueprint (ICLR 2025 · similarity 0.82)
- WorldCoder-Bench: Benchmarking Physically Grounded 3D World Synthesis (NeurIPS 2026 · similarity 0.82)
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.81)
- ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows (ICLR 2026 · similarity 0.81)
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.
