# AstraLOD3: Zero-shot multimodal agentic reconstruction of LOD3 building models

arXiv: https://arxiv.org/abs/2609.28061 · HTML: https://mlsanity.com/paper/2609.28061/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use.
- Builds on 17 of its 49 references that are in the map.
- First posted on arXiv 23 Sep 2026.

## Placed in

- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 17 of 17)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [Code as Policies: Language Model Programs for Embodied Control](https://mlsanity.com/paper/2209.07753/) (2022)
- [A Point Set Generation Network for 3D Object Reconstruction from a Single Image](https://mlsanity.com/paper/1612.00603/) (2016)
- [LRM: Large Reconstruction Model for Single Image to 3D](https://mlsanity.com/paper/2311.04400/) (2023)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents](https://mlsanity.com/paper/2311.05437/) (2023)
- [SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Code](https://mlsanity.com/paper/2403.01248/) (2024)
- [City3D: Large-Scale Building Reconstruction from Airborne LiDAR Point Clouds](https://mlsanity.com/paper/2201.10276/) (2022)
- [Reconstructing Compact Building Models from Point Clouds Using Deep Implicit Fields](https://mlsanity.com/paper/2112.13142/) (2021)
- [MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds](https://mlsanity.com/paper/2508.14879/) (2025)
- [Scan2LoD3: Reconstructing semantic 3D building models at LoD3 using ray casting and Bayesian networks](https://mlsanity.com/paper/2305.06314/) (2023)
- [BIMgent: Towards Autonomous Building Modeling via Computer-use Agents](https://mlsanity.com/paper/2506.07217/) (2025)
- [SVI2LoD3: Agent-Driven Reconstruction of LoD3 Facade Openings in Semantic 3D City Models from Volunteered Street View Imagery using Large Language and Visual Models](https://mlsanity.com/paper/2608.29992/) (2026)

## Abstract (arXiv)

Automated LOD3 building modeling typically relies on purpose-built geometric or learning-based pipelines, limiting flexibility across heterogeneous buildings and input evidence conditions. This study investigates whether Astra, a general-purpose multimodal foundation model, can address these limitations through zero-shot reconstruction of LOD3 building models within an agentic framework under bounded autonomy. AstraLOD3 combines multi-view images, calibrated cameras, and a filtered sparse SfM point cloud with a natural-language reconstruction specification, while the Astra agent dynamically selects and executes computational procedures using Python and Blender. Across 35 runs, including 24 benchmark buildings, AstraLOD3 achieved a mean FRDS of 0.9647 and geometric agreement comparable to that of previous purpose-built methods. Controlled ablations further revealed the effects of reconstruction guidance, evidence modalities, model configuration, and run-to-run variability. The results demonstrate that structured LOD3 reconstruction can be formulated as a constrained agentic process rather than as a fixed pipeline. Future work will investigate adaptive refinement, user-guided correction, task-specific specialization, and damage-aware reconstruction.