研究人员推出CAIRN，这是一种新颖的拓扑感知大型多模态模型，用于理解复杂的多房间三维场景。与以往仅限于单房间的模型不同，CAIRN明确地推理物体关系和房间连通性。它通过集成图神经网络和学习到的房间令牌来实现这一点，从而实现尊重场景拓扑的层次化注意力。CAIRN在新推出的CAIRN-MR基准上进行了评估，在多房间任务上展示了比现有3D-LLM显著的性能提升。 AI
影响 该模型提升了多模态模型理解复杂、真实世界三维环境的能力，可能对机器人和虚拟现实应用产生影响。
排序理由 该集群描述了一篇介绍新模型和基准的研究论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2607.06534v1 Announce Type: new Abstract: Existing 3D scene-grounded Large Language Models (3D-LLMs) focus on answering questions grounded in simplified single-room 3D scenes, lacking the ability to reason over real-world household environments containing multiple interconn…
Existing 3D scene-grounded Large Language Models (3D-LLMs) focus on answering questions grounded in simplified single-room 3D scenes, lacking the ability to reason over real-world household environments containing multiple interconnected rooms and diverse object categories. We in…