DiSR：解耦3D建模与空间推理的新框架

AI 摘要

该研究提出了一种名为DiSR的框架，将3D感知与空间推理分离，利用现成的专家感知模型重建结构化3D证据，并通过LoRA微调LLM仅基于这些几何证据进行推理。DiSR在空间推理基准上取得了有竞争力的性能，同时提供了更好的可解释性、模块化和计算效率，表明显式分离感知与推理是端到端建模的可扩展替代方案。

正文节选

Computer Science > Machine Learning Title:Disentangling 3D Modeling from Spatial Reasoning View PDF HTML (experimental) Abstract:In this work, we explore an alternative paradigm for spatial reasoning by explicitly disentangling 3D perception from reasoning, rather than jointly acquiring implicit 3D perception and reasoning through large-scale training. Our key observation is that modern perception models excel at estimating continuous 3D geometry, whereas large language models (LLMs)

发布时间：2026-08-07 12:00

抓取时间：2026-08-07 14:19

来源机构：arXiv