AINeutralarXiv – CS AI · Mar 37/109
🧠
Evaluating Theory of Mind and Internal Beliefs in LLM-Based Multi-Agent Systems
Researchers introduce a novel multi-agent AI architecture that integrates Theory of Mind, internal beliefs, and symbolic solvers to improve collaborative decision-making in LLM-based systems. The study evaluates this architecture across different language models in resource allocation scenarios, revealing complex interactions between LLM capabilities and cognitive mechanisms.