BuildingGPT：基于强化学习的自回归建筑线框重建模型
BuildingGPT: Auto-Regressive Building Wireframe Reconstruction Model with Reinforcement Learning
把建筑线框重建从检测/扩散范式改写为「点云→潜变量→分层词元序列」的自回归下一词元预测，并用 DPO 做后训练对齐；做点云结构化重建或想把 LLM 式序列生成+偏好优化搬到几何任务的人可直接借鉴这套词元化与两阶段训练流程。
原文依据 reformulates the building wireframe reconstruction task into a sequence prediction problem
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 BuildingGPT，一种利用强化学习从点云重建建筑线框的新型自回归模型。与基于检测或扩散模型的先前工作不同，BuildingGPT 将建筑线框重建任务重构为序列预测问题。基于分层式的建筑线框词元化，线框序列以结构感知与语义感知的顺序组织，用于下一词元预测。点云编码器首先将输入点云转换为固定长度的潜变量编码，作为序列的起始。随后，BuildingGPT 以该潜变量编码和先前生成的词元为条件，自回归地预测词元。在预测出词元序列后，通过去词元化得到建筑线框。为提升模型性能，我们采用包含预训练与后训练的两阶段训练范式。在自回归预训练之后，采用直接偏好优化（DPO）作为后训练策略，使重建结果与人类偏好对齐。在大规模 MunichWF 数据集上的大量实验表明，BuildingGPT 优于现有的最优方法。我们承诺公开代码与数据集。
英文原文摘要 1291 字符
In this paper, we propose BuildingGPT, a novel auto-regressive model for building wireframe reconstruction from point clouds with reinforcement learning.Unlike prior works based on detection or diffusion models, BuildingGPT reformulates the building wireframe reconstruction task into a sequence prediction problem.Based on the hierarchical building wireframe tokenization, the wireframe sequences are organized in a structurally- and semantically-aware order for the next-token prediction.The point cloud encoder first transforms the input point cloud into a fixed-length latent code that serves as the starting of the sequence.Then, BuildingGPT auto-regressively predicts tokens conditioned on the latent code and previously generated tokens.With token sequence predicted, the building wireframe is obtained through detokenization.To enhance the model performance, we adopt a two-stage training paradigm including the pre-training and post-training.After the auto-regressive pre-training, Direct Preference Optimization (DPO) is employed as a post-training strategy to align reconstruction results with human preferences.Extensive experiments on the large-scale MunichWF dataset show that BuildingGPT outperforms existing state-of-the-art methods.We commit to release the code and dataset.