Learning Dynamic Belief Graphs for Theory-of-mind Reasoning
On the map: LLM reasoning → Reasoning benchmarks and evaluation
Abstract (arXiv)
Theory of Mind (ToM) reasoning with Large Language Models (LLMs) requires inferring how people's implicit, evolving beliefs shape what they seek and how they act under uncertainty -- especially in high-stakes settings such as disaster response, emergency medicine, and human-in-the-loop autonomy. Prior approaches either prompt LLMs directly or use latent-state models that treat beliefs as static and independent, often producing incoherent mental models over time and weak reasoning in dynamic contexts. We introduce a structured cognitive trajectory model for LLM-based ToM that represents mental state as a dynamic belief graph, jointly inferring latent beliefs, learning their time-varying dependencies, and linking belief evolution to information seeking and decisions. Our model contributes (i) a novel projection from textualized probabilistic statements to consistent probabilistic graphical model updates, (ii) an energy-based factor graph representation of belief interdependencies, and (iii) an ELBO-based objective that captures belief accumulation and delayed decisions. Across multiple real-world disaster evacuation datasets, our model significantly improves action prediction and recovers interpretable belief trajectories consistent with human reasoning, providing a principled module for augmenting LLMs with ToM in high-uncertainty environment. https://anonymous.4open.science/r/ICML_submission-6373/
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 12 of its 57 references that are in the map. First posted on arXiv 20 Mar 2026.
Builds on
12 of 57 references.
Closest ancestors · 1
Same direction, newest first
Other references · 11
- 2017Attention Is All You Needcited by 59,932
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2013Auto-Encoding Variational Bayescited by 3,980
- 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
- 2023Large Language Models Can Be Easily Distracted by Irrelevant Contextcited by 820
- 2016Structured Inference Networks for Nonlinear State Space Modelscited by 182
- 2024Personalization of Large Language Models: A Surveycited by 100
- 2015Structured Prediction Energy Networkscited by 59
- 2023Factor Graph Neural Networkscited by 13
- 2024EgoSocialArena: Benchmarking the Social Intelligence of Large Language Models from a First-person Perspectivecited by 10
- 2025Large Language Models as Theory of Mind Aware Generative Agents with Counterfactual Reflectioncited by 10
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Dynamic Theory of Mind as a Temporal Memory Problem: Evidence from Large Language Models
- 2025Overcoming Multi-step Complexity in Multimodal Theory-of-Mind Reasoning: A Scalable Bayesian Planner
- 2026PDDL-Mind: Large Language Models are Capable on Belief Reasoning with Reliable State Tracking
- 2025Hypothesis-Driven Theory-of-Mind Reasoning for Large Language Models
- 2025Do Theory of Mind Benchmarks Need Explicit Human-like Reasoning in Language Models?
- 2026Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models
- 2026Reasoning aligns language models to human cognition
- 2026To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks