Research map / Large language models / Instruction and fine-tuning
BrepLLM: Enabling Large Language Models to Understand Boundary Representations
Liyuan Deng, Hao Guo, Yongkang Dai, Yunpeng Bai, Yifan Zhu, Yuanyuan Gao, Huaxi Huang, Yilei Shi
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 |
|---|---|---|---|
| feed native B-rep via BrepEncoder instead of point clouds or rendered images | BrepLLM · Brep2Text captioning | Qwen3-30B score: 56.58 -> 61.36 (vs MiniGPT-3D) | Improved |
| feed native B-rep to LLM | BrepLLM · Brep2Text classification | average accuracy: 56.25 -> 60.0 | Improved |
| B-rep input vs point cloud input (human eval) | BrepLLM · Brep2Text captioning | hallucination count: 0.86 -> 0.83 | Improved |
| add adaptive UV sampling (ablation) | BrepLLM · Brep2Text captioning | Qwen3-30B score: 59.21 -> 61.36 | Improved |
| add hierarchical BrepEncoder (ablation) | BrepLLM · Brep2Text captioning | Qwen3-30B score: 57.39 -> 61.36 | Improved |
| add Stage II (LoRA 3D-language alignment) after Stage I (ablation) | BrepLLM · Brep2Text captioning | Qwen3-30B score: 43.83 -> 61.36 | Improved |
| add Stage I (MLP bridging) before Stage II (ablation) | BrepLLM · Brep2Text captioning | Qwen3-30B score: 59.12 -> 61.36 | Improved |
Closest papers by meaning
- 3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models (NeurIPS 2026 · similarity 0.87)
- Merge3D: Efficient 3D Multimodal LLMs via Joint 2D-3D Token Merging (CVPR 2026 · similarity 0.86)
- ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding (CVPR 2024 · similarity 0.86)
- ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding (NeurIPS 2025 · similarity 0.86)
- More Text, Less Point: Towards 3D Data-Efficient Point-Language Understanding (AAAI 2025 · similarity 0.86)
- BrepGiff: Lightweight Generation of Complex B-rep with 3D GAT Diffusion (CVPR 2025 · 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.
