Lune

EMNLP Findings2025

INDOORWORLD : Integrating Physical Task Solving and Social Simulation in A Heterogeneous Multi-Agent Environment

Dekun Wu, Frederik Brudy, Bang Liu, Yi Wang

Abstract

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 multiagent collaboration, resource competition, and spatial layout on agent behavior.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Reflexion: language agents with verbal reinforcement learning cites ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
  • Reflexion: language agents with verbal reinforcement learning cites Generative Agents: Interactive Simulacra of Human Behavior
  • Building Cooperative Embodied Agents Modularly with Large Language Models cites Generative Agents: Interactive Simulacra of Human Behavior
  • MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning cites Generative Agents: Interactive Simulacra of Human Behavior
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on17

  1. Reflexion: language agents with verbal reinforcement learningNoah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan et al.NeurIPS 2023 · 5,828 citations
  2. Generative Agents: Interactive Simulacra of Human BehaviorJoon Sung Park, Joseph C. O'Brien, Carrie Jun Cai, Meredith Ringel Morris et al.UIST 2023 · 1,882 citations
  3. ALFWorld: Aligning Text and Embodied Environments for Interactive LearningMohit Shridhar, Xingdi Yuan, Marc-Alexandre Côté, Yonatan Bisk et al.ICLR 2021 · 819 citations
  4. Building Cooperative Embodied Agents Modularly with Large Language ModelsHongxin Zhang, Weihua Du, Jiaming Shan, Qinhong Zhou et al.ICLR 2024 · 303 citations
  5. MedAgents: Large Language Models as Collaborators for Zero-shot Medical ReasoningXiangru Tang, Anni Zou, Zhuosheng Zhang, Ziming Li et al.ACL Findings 2024 · 160 citations

Related papers

  1. SimWorld: An Open-ended Simulator for Agents in Physical and Social WorldsXiaokang Ye, Jiawei Ren, Yan Zhuang, Xuhong He et al.NeurIPS 2025 · 8 citations
  2. Your Co-Workers Matter: Evaluating Collaborative Capabilities of Language Models in Blocks WorldGuande Wu, Chen Zhao, Cláudio T. Silva, He HeACL Findings 2024 · 3 citations
  3. LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon GameplayYihuai Lan, Zhiqiang Hu, Lei Wang, Yang Wang et al.EMNLP 2024 · 10 citations
  4. RoboTHOR: An Open Simulation-to-Real Embodied AI PlatformMatt Deitke, Winson Han, Alvaro Herrasti, Aniruddha Kembhavi et al.CVPR 2020
  5. Code2Worlds: Empowering Coding LLMs for 4D World GenerationYi Zhang, Yunshuang Wang, Zeyu Zhang, Hao TangICML 2026 · 6 citations
Dusk over the sea between two cliffs drawn in fine vertical lines