AutoToM: Scaling Model-based Mental Inference via Automated Agent Modeling
This work addresses the challenge of developing socially intelligent agents by improving ToM reasoning, though it appears incremental as it builds on model-based inference with LLM integration.
The authors tackled the problem of Theory of Mind (ToM) reasoning by introducing AutoToM, an automated agent modeling method that outperforms existing approaches across five diverse benchmarks, achieving scalable, robust, and interpretable mental inference.
Theory of Mind (ToM), the ability to understand people's minds based on their behavior, is key to developing socially intelligent agents. Current approaches to ToM reasoning either rely on prompting Large Language Models (LLMs), which are prone to systematic errors, or use handcrafted, rigid agent models for model-based inference, which are more robust but fail to generalize across domains. In this work, we introduce AutoToM, an automated agent modeling method for scalable, robust, and interpretable mental inference. Given a ToM problem, AutoToM first proposes an initial agent model and then performs automated Bayesian inverse planning based on this model, leveraging an LLM backend. Guided by inference uncertainty, it iteratively refines the model by introducing additional mental variables and/or incorporating more timesteps in the context. Across five diverse benchmarks, AutoToM outperforms existing ToM methods and even large reasoning models. Additionally, we show that AutoToM can produce human-like confidence estimates and enable online mental inference for embodied decision-making.