IndoorWorld: Integrating Physical Task Solving and Social Simulation in A Heterogeneous Multi-Agent Environment
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916793959841792 |
|---|---|
| author | Wu, Dekun Brudy, Frederik Liu, Bang Wang, Yi |
| author_facet | Wu, Dekun Brudy, Frederik Liu, Bang Wang, Yi |
| contents | Virtual environments are essential to AI agent research. Existing environments for LLM agent research typically focus on either physical task solving or social simulation, with the former oversimplifying agent individuality and social dynamics, and the latter lacking physical grounding of social behaviors. We introduce IndoorWorld, a heterogeneous multi-agent environment that tightly integrates physical and social dynamics. By introducing novel challenges for LLM-driven agents in orchestrating social dynamics to influence physical environments and anchoring social interactions within world states, IndoorWorld opens up possibilities of LLM-based building occupant simulation for architectural design. We demonstrate the potential with a series of experiments within an office setting to examine the impact of multi-agent collaboration, resource competition, and spatial layout on agent behavior. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_12331 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | IndoorWorld: Integrating Physical Task Solving and Social Simulation in A Heterogeneous Multi-Agent Environment Wu, Dekun Brudy, Frederik Liu, Bang Wang, Yi Multiagent Systems Artificial Intelligence Virtual environments are essential to AI agent research. Existing environments for LLM agent research typically focus on either physical task solving or social simulation, with the former oversimplifying agent individuality and social dynamics, and the latter lacking physical grounding of social behaviors. We introduce IndoorWorld, a heterogeneous multi-agent environment that tightly integrates physical and social dynamics. By introducing novel challenges for LLM-driven agents in orchestrating social dynamics to influence physical environments and anchoring social interactions within world states, IndoorWorld opens up possibilities of LLM-based building occupant simulation for architectural design. We demonstrate the potential with a series of experiments within an office setting to examine the impact of multi-agent collaboration, resource competition, and spatial layout on agent behavior. |
| title | IndoorWorld: Integrating Physical Task Solving and Social Simulation in A Heterogeneous Multi-Agent Environment |
| topic | Multiagent Systems Artificial Intelligence |
| url | https://arxiv.org/abs/2506.12331 |