Enregistré dans:
Détails bibliographiques
Auteurs principaux: Jafari, Mehdi, Hua, Devin Yuncheng, Xue, Hao, Salim, Flora
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2502.14171
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909616707731456
author Jafari, Mehdi
Hua, Devin Yuncheng
Xue, Hao
Salim, Flora
author_facet Jafari, Mehdi
Hua, Devin Yuncheng
Xue, Hao
Salim, Flora
contents Natural language interaction with agentic Artificial Intelligence (AI), driven by Large Language Models (LLMs), is expected to remain a dominant paradigm in the near future. While humans instinctively align their communication with mental states -- an ability known as Theory of Mind (ToM), current LLM powered systems exhibit significant limitations in this regard. This study examines the extent to which open source language models (LLaMA) can capture and preserve ToM related information and how effectively it contributes to consistent ToM reasoning in generated responses. We further investigate whether explicit manipulation of ToM related components, such as beliefs, desires, and intentions, can enhance response alignment. Experiments on two LLaMA 3 variants demonstrate that incorporating ToM informed alignment improves response quality, achieving win rates of 67 and 63 percent for the 3B and 8B models, respectively. These findings highlight the potential of ToM driven strategies to improve alignment in LLM based conversational agents.
format Preprint
id arxiv_https___arxiv_org_abs_2502_14171
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhancing Conversational Agents with Theory of Mind: Aligning Beliefs, Desires, and Intentions for Human-Like Interaction
Jafari, Mehdi
Hua, Devin Yuncheng
Xue, Hao
Salim, Flora
Computation and Language
Natural language interaction with agentic Artificial Intelligence (AI), driven by Large Language Models (LLMs), is expected to remain a dominant paradigm in the near future. While humans instinctively align their communication with mental states -- an ability known as Theory of Mind (ToM), current LLM powered systems exhibit significant limitations in this regard. This study examines the extent to which open source language models (LLaMA) can capture and preserve ToM related information and how effectively it contributes to consistent ToM reasoning in generated responses. We further investigate whether explicit manipulation of ToM related components, such as beliefs, desires, and intentions, can enhance response alignment. Experiments on two LLaMA 3 variants demonstrate that incorporating ToM informed alignment improves response quality, achieving win rates of 67 and 63 percent for the 3B and 8B models, respectively. These findings highlight the potential of ToM driven strategies to improve alignment in LLM based conversational agents.
title Enhancing Conversational Agents with Theory of Mind: Aligning Beliefs, Desires, and Intentions for Human-Like Interaction
topic Computation and Language
url https://arxiv.org/abs/2502.14171