Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hashimoto, Saki, Hasegawa, Shoichi, Ishikawa, Tomochika, Taniguchi, Akira, Hagiwara, Yoshinobu, Hafi, Lotfi El, Taniguchi, Tadahiro
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918142071013376
author Hashimoto, Saki
Hasegawa, Shoichi
Ishikawa, Tomochika
Taniguchi, Akira
Hagiwara, Yoshinobu
Hafi, Lotfi El
Taniguchi, Tadahiro
author_facet Hashimoto, Saki
Hasegawa, Shoichi
Ishikawa, Tomochika
Taniguchi, Akira
Hagiwara, Yoshinobu
Hafi, Lotfi El
Taniguchi, Tadahiro
contents Robots operating in domestic and office environments must understand object ownership to correctly execute instructions such as ``Bring me my cup.'' However, ownership cannot be reliably inferred from visual features alone. To address this gap, we propose Active Ownership Learning (ActOwL), a framework that enables robots to actively generate and ask ownership-related questions to users. ActOwL employs a probabilistic generative model to select questions that maximize information gain, thereby acquiring ownership knowledge efficiently to improve learning efficiency. Additionally, by leveraging commonsense knowledge from Large Language Models (LLM), objects are pre-classified as either shared or owned, and only owned objects are targeted for questioning. Through experiments in a simulated home environment and a real-world laboratory setting, ActOwL achieved significantly higher ownership clustering accuracy with fewer questions than baseline methods. These findings demonstrate the effectiveness of combining active inference with LLM-guided commonsense reasoning, advancing the capability of robots to acquire ownership knowledge for practical and socially appropriate task execution.
format Preprint
id arxiv_https___arxiv_org_abs_2509_12754
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model
Hashimoto, Saki
Hasegawa, Shoichi
Ishikawa, Tomochika
Taniguchi, Akira
Hagiwara, Yoshinobu
Hafi, Lotfi El
Taniguchi, Tadahiro
Robotics
Artificial Intelligence
Human-Computer Interaction
Machine Learning
Robots operating in domestic and office environments must understand object ownership to correctly execute instructions such as ``Bring me my cup.'' However, ownership cannot be reliably inferred from visual features alone. To address this gap, we propose Active Ownership Learning (ActOwL), a framework that enables robots to actively generate and ask ownership-related questions to users. ActOwL employs a probabilistic generative model to select questions that maximize information gain, thereby acquiring ownership knowledge efficiently to improve learning efficiency. Additionally, by leveraging commonsense knowledge from Large Language Models (LLM), objects are pre-classified as either shared or owned, and only owned objects are targeted for questioning. Through experiments in a simulated home environment and a real-world laboratory setting, ActOwL achieved significantly higher ownership clustering accuracy with fewer questions than baseline methods. These findings demonstrate the effectiveness of combining active inference with LLM-guided commonsense reasoning, advancing the capability of robots to acquire ownership knowledge for practical and socially appropriate task execution.
title Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model
topic Robotics
Artificial Intelligence
Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2509.12754