Paper 2607.27201
Mental World Modeling
- Published
- Jul 2026
- Research lab
- Independent
- Citations
- 0
- GitHub
- 16 stars
01 In brief
Summary
The paper introduces Mental World Modeling (MWM), a framework for world models that jointly represent physical and mental dynamics to predict human decisions.
MWM maintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions update both components.
The authors implement MENTIS, a training-free, inspectable baseline that parses scenes into structured states, generates target observations, decomposes actions, simulates coupled transitions, and evaluates branches.
On the Menti-Bench dataset (448 records across text, image, and video), experiments with 8 LLM-based world models show that explicit mental modeling is necessary: full MWM achieves 87.9% F1 on average, outperforming direct answering (63.3%) and self-consistency (77.9%).
Ablations removing mental or physical channels degrade performance by 12.1 and 16.5 points respectively.
Oracle interventions identify transition simulation as the main bottleneck, accounting for 3.5 of the 7.8-point gap to human performance (98.5%).
Gains are largest on interpersonal scenes (+26.4 points over direct answering) and consistent across modalities.
The authors conclude that MWM represents a next stage of world modeling, from simulating physical scenes to simulating the minds that act in them.
02 From the paper
Abstract
World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model rather than posthoc rationales: MWM aintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions jointly update both components. We instantiate the framework in MENTIS, a training-free and fully inspectable baseline that decomposes the process into state parsing, target-observation generation, action decomposition, coupled physical and mental transition, and branch-level value evaluation. On a manually constructed, quality-controlled dataset of situated decision scenarios spanning text, image, and sounding-video stories, experiments with 8 modern LLM-based world models demonstrate that explicitly modeling the mental state is essential for predicting human decisions. Deeper analyses further expose the bottlenecks of current mental world modeling. We expect MWM as a next stage of world modeling, from simulating physical scenes to simulating the minds that act in them.