Mental World Model 框架提出：让 AI 建模主观心理状态的演化
Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability
一篇关于 Theory of Mind 的论文，提出的 MMEA 在 DynToM 上拿到 96.48%，比主流 LLM 高出不少，做心理建模方向的可以看看。
arXiv 论文提出 Mental World Model（MWM）及 Multi-Branch Mental Evolution Architecture（MMEA）实现，显式构建主观世界并模拟心理状态的持续演化。在 DynToM 基准上 MMEA 达到 96.48% 准确率，在 RecToM 上达到 94.00%，超过主流 LLM 和专门的 Theory of Mind 方法。在心理状态与行为预测、长序列预测、隐式心理状态推理三项任务上分别提升 28.7、39.3 和 16.6 个百分点。现有 LLM 主要从语言信息推断心理状态，该框架补充了历史依赖和认知-心理交互机制。
Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability
Understanding and simulating how subjective cognition and mental states emerge, evolve, and shape behavior is essential for advancing artificial intelligence toward deeper social cognition and general intelligence. Existing large language models (LLMs) and Theory of Mind (ToM) methods primarily infer mental states from linguistic information, lacking unified mechanisms for subjective world construction, cognitive-psychological interactions, dynamic evolution, and historical dependencies. This paper introduces the Mental World Model (MWM) and its implementation, the Multi-Branch Mental Evolution Architecture (MMEA), which explicitly constructs subjective worlds, models interactions between cognition and mental states, and simulates continuous mental evolution shaped by historical experiences. On DynToM and RecToM, MMEA achieves 96.48% and 94.00% accuracy, respectively, outperforming leading LLMs and specialized ToM methods. Additional experiments demonstrate improvements of 28.7, 39.3, and 16.6 percentage points in mental state and behavior prediction, long-sequence prediction, and implicit mental state inference, respectively. The results demonstrate that MMEA provides an interpretable framework for continuous mental modeling, enabling AI to move beyond surface-level semantic representations to infer individuals' underlying mental states from linguistic information and dynamically simulate their internal cognitive processes and evolution over time. MWM also provides a promising foundation for causal reasoning, personalized AI, sustained human-AI interaction, and future advances toward artificial general intelligence (AGI).
