Research map / 3D vision / 3D generation
LAM: Language Articulated Object Modelers
Yipeng Gao, Yunhao Ge, Peilin Cai, Daniel Seita, Laurent Itti
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 Geometry Checker (ablation) | LAM · 83 captions | accuracy (%): 50.6 -> 61.4 | Improved |
| add Articulation Checker (ablation) | LAM · 83 captions | accuracy (%): 50.6 -> 56.6 | Improved |
| add both Checkers (1 iteration) (ablation) | LAM · 83 captions | accuracy (%): 50.6 -> 66.3 | Improved |
| increase refinement iterations from 1 to 3 (ablation) | LAM · 83 captions | accuracy (%): 66.3 -> 75.9 | Improved |
| use multi-view images for 2D Checker (ablation) | LAM · 83 captions | accuracy (%): 60.2 -> 65.1 | Improved |
| add image sequence of motion states for Articulation Checker (ablation) | LAM · 83 captions | accuracy (%): 65.1 -> 71.1 | Improved |
| use 3D checker only instead of 2D (ablation) | LAM · 83 captions | accuracy (%): 71.1 -> 62.7 | Hurt |
| combine 2D and 3D Checkers (ablation) | LAM · 83 captions | accuracy (%): 71.1 -> 75.9 | Improved |
| swap Geometry Coder to Claude-4.1-Opus (ablation) | LAM · 83 captions | accuracy (%): 75.9 -> 80.7 | Improved |
| use weaker Link Designer and Geometry Coder (Gemini-2.5-flash) (ablation) | LAM · 83 captions | accuracy (%): 75.9 -> 57.9 | Hurt |
| finetune Qwen3-VL-8B on LAMBench | LAM · Part-Mobility five classes | success rate (%): 36.8 -> 51.6 | Improved |
| finetune Qwen3-VL-8B on LAMBench | LAM · Part-Mobility general classes | success rate (%): 44.3 -> 49.6 | 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.
- Real2Code — Real2Code: Reconstruct Articulated Objects via Code Generation (ICLR 2025) · 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
- ArtFormer: Controllable Generation of Diverse 3D Articulated Objects (CVPR 2025 · similarity 0.88)
- 3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code (NeurIPS 2026 · similarity 0.87)
- Let Language Constrain Geometry: Vision–Language Models as Semantic and Spatial Critics for 3D Generation (ICML 2026 · similarity 0.87)
- MeshArt: Generating Articulated Meshes with Structure-Guided Transformers (CVPR 2025 · similarity 0.87)
- Articulate-Anything: Automatic Modeling of Articulated Objects via a Vision-Language Foundation Model (ICLR 2025 · similarity 0.87)
- Socratic-Geo: Synthetic Data Generation and Cross-Modal Geometric Reasoning via Multi-Agent Interaction (CVPR 2026 · similarity 0.86)
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.
