Adaptive Theory of Mind for LLM-based Multi-Agent Coordination

• URL: http://arxiv.org/abs/2603.16264v1
• Date: Tue, 17 Mar 2026 08:52:57 GMT
• Title: Adaptive Theory of Mind for LLM-based Multi-Agent Coordination
• Abstract summary: Theory of Mind (ToM) refers to the ability to reason about others' mental states.<n>We design an adaptive ToM (A-ToM) agent, which can align in ToM orders with its partner.
• Score: 17.374096437201413
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Theory of Mind (ToM) refers to the ability to reason about others' mental states, and higher-order ToM involves considering that others also possess their own ToM. Equipping large language model (LLM)-driven agents with ToM has long been considered to improve their coordination in multiagent collaborative tasks. However, we find that misaligned ToM orders-mismatches in the depth of ToM reasoning between agents-can lead to insufficient or excessive reasoning about others, thereby impairing their coordination. To address this issue, we design an adaptive ToM (A-ToM) agent, which can align in ToM orders with its partner. Based on prior interactions, the agent estimates the partner's likely ToM order and leverages this estimation to predict the partner's action, thereby facilitating behavioral coordination. We conduct empirical evaluations on four multi-agent coordination tasks: a repeated matrix game, two grid navigation tasks and an Overcooked task. The results validate our findings on ToM alignment and demonstrate the effectiveness of our A-ToM agent. Furthermore, we discuss the generalizability of our A-ToM to non-LLM-based agents, as well as what would diminish the importance of ToM alignment.

Related papers

• Enhancing Decision-Making with Large Language Models through Multi-Agent Fictitious Play [53.56274149236814]
  Multi-Agent Fictitious Play (MAFP) is a novel MAS paradigm that represents stakeholder stances as agents.<n>MAFP iteratively updates each agent's decision by best responding to the empirical mixture of other agents' past decisions.<n>We evaluate MAFP on challenging decision-making tasks that test the capability of deciding strategies for competitive scenarios prior to acting.
  arXiv  Detail & Related papers  (2026-06-17T17:31:06Z)
• Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models [10.629439705877054]
  We study whether large language models (LLMs) exhibit genuine Theory of Mind (ToM) capabilities.<n>We introduce a handcrafted, richly annotated ToM dataset, including classic and perturbed false belief tasks.<n>We show a steep drop in ToM capabilities under task perturbation for all evaluated LLMs, questioning the notion of any robust form of ToM being present.
  arXiv  Detail & Related papers  (2026-02-25T16:24:35Z)
• LLM-Hanabi: Evaluating Multi-Agent Gameplays with Theory-of-Mind and Rationale Inference in Imperfect Information Collaboration Game [47.019077016616144]
  This study introduces LLM-Hanabi, a novel benchmark that uses the cooperative game Hanabi to evaluate the rationale inference and ToM.<n>Across a range of models, we find a significant positive correlation between ToM and in-game success.<n>We conclude that prioritizing first-order ToM is a promising direction for enhancing the collaborative capabilities of future models.
  arXiv  Detail & Related papers  (2025-10-06T16:17:24Z)
• Theory of Mind Using Active Inference: A Framework for Multi-Agent Cooperation [4.033462754517733]
  Theory of Mind (ToM) enables agents to reason about others' beliefs while planning their own actions.<n>We present a novel approach to multi-agent cooperation by implementing ToM within active inference.
  arXiv  Detail & Related papers  (2025-08-01T08:02:35Z)
• SAND: Boosting LLM Agents with Self-Taught Action Deliberation [54.48979740613828]
  Large Language Model (LLM) agents are commonly tuned with supervised finetuning on ReAct-style expert trajectories or preference optimization over pairwise rollouts.<n>We propose Self-taught ActioN Deliberation (SAND) framework, enabling LLM agents to explicitly deliberate over candidate actions before committing to one.<n>SAND achieves an average 20% improvement over initial supervised finetuning and also outperforms state-of-the-art agent tuning approaches.
  arXiv  Detail & Related papers  (2025-07-10T05:38:15Z)
• Large Language Models as Theory of Mind Aware Generative Agents with Counterfactual Reflection [31.38516078163367]
  ToM-agent is designed to empower LLMs-based generative agents to simulate ToM in open-domain conversational interactions.<n>ToM-agent disentangles the confidence from mental states, facilitating the emulation of an agent's perception of its counterpart's mental states.<n>Our findings indicate that the ToM-agent can grasp the underlying reasons for their counterpart's behaviors beyond mere semantic-emotional supporting or decision-making based on common sense.
  arXiv  Detail & Related papers  (2025-01-26T00:32:38Z)
• Mutual Theory of Mind in Human-AI Collaboration: An Empirical Study with LLM-driven AI Agents in a Real-time Shared Workspace Task [56.92961847155029]
  Theory of Mind (ToM) significantly impacts human collaboration and communication as a crucial capability to understand others. Mutual Theory of Mind (MToM) arises when AI agents with ToM capability collaborate with humans. We find that the agent's ToM capability does not significantly impact team performance but enhances human understanding of the agent.
  arXiv  Detail & Related papers  (2024-09-13T13:19:48Z)
• Emergence of Theory of Mind Collaboration in Multiagent Systems [65.97255691640561]
  We propose an adaptive training algorithm to develop effective collaboration between agents with ToM. We evaluate our algorithms with two games, where our algorithm surpasses all previous decentralized execution algorithms without modeling ToM.
  arXiv  Detail & Related papers  (2021-09-30T23:28:00Z)
• UneVEn: Universal Value Exploration for Multi-Agent Reinforcement Learning [53.73686229912562]
  We propose a novel MARL approach called Universal Value Exploration (UneVEn) UneVEn learns a set of related tasks simultaneously with a linear decomposition of universal successor features. Empirical results on a set of exploration games, challenging cooperative predator-prey tasks requiring significant coordination among agents, and StarCraft II micromanagement benchmarks show that UneVEn can solve tasks where other state-of-the-art MARL methods fail.
  arXiv  Detail & Related papers  (2020-10-06T19:08:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.