Saved in:
Bibliographic Details
Main Authors: Sagara, Rikunari, Terao, Koichiro, Iwahashi, Naoto
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.12321
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913845250883584
author Sagara, Rikunari
Terao, Koichiro
Iwahashi, Naoto
author_facet Sagara, Rikunari
Terao, Koichiro
Iwahashi, Naoto
contents This paper introduces an open-source simulator, BeliefNest, designed to enable embodied agents to perform collaborative tasks by leveraging Theory of Mind. BeliefNest dynamically and hierarchically constructs simulators within a Minecraft environment, allowing agents to explicitly represent nested belief states about themselves and others. This enables agent control in open-domain tasks that require Theory of Mind reasoning. The simulator provides a prompt generation mechanism based on each belief state, facilitating the design and evaluation of methods for agent control utilizing large language models (LLMs). We demonstrate through experiments that agents can infer others' beliefs and predict their belief-based actions in false-belief tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12321
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BeliefNest: A Joint Action Simulator for Embodied Agents with Theory of Mind
Sagara, Rikunari
Terao, Koichiro
Iwahashi, Naoto
Artificial Intelligence
This paper introduces an open-source simulator, BeliefNest, designed to enable embodied agents to perform collaborative tasks by leveraging Theory of Mind. BeliefNest dynamically and hierarchically constructs simulators within a Minecraft environment, allowing agents to explicitly represent nested belief states about themselves and others. This enables agent control in open-domain tasks that require Theory of Mind reasoning. The simulator provides a prompt generation mechanism based on each belief state, facilitating the design and evaluation of methods for agent control utilizing large language models (LLMs). We demonstrate through experiments that agents can infer others' beliefs and predict their belief-based actions in false-belief tasks.
title BeliefNest: A Joint Action Simulator for Embodied Agents with Theory of Mind
topic Artificial Intelligence
url https://arxiv.org/abs/2505.12321