MindGames: Targeting Theory of Mind in Large Language Models with Dynamic Epistemic Modal Logic
This addresses the challenge of evaluating ToM, a critical component of intelligence, in AI models, with incremental improvements in assessment methods.
The paper tackled the problem of assessing Theory of Mind (ToM) in large language models by using dynamic epistemic logic to generate controlled problems and new verbalization techniques, finding that scaling from 70M to 6B and 350M to 174B parameters does not consistently improve results beyond random chance, though GPT-4 showed superior capabilities.
Theory of Mind (ToM) is a critical component of intelligence but its assessment remains the subject of heated debates. Prior research applied human ToM assessments to natural language processing models using either human-created standardized tests or rule-based templates. However, these methods primarily focus on simplistic reasoning and require further validation. Here, we leverage dynamic epistemic logic to isolate a particular component of ToM and to generate controlled problems. We also introduce new verbalization techniques to express these problems in English natural language. Our findings indicate that some language model scaling (from 70M to 6B and 350M to 174B) does not consistently yield results better than random chance. While GPT-4 demonstrates superior epistemic reasoning capabilities, there is still room for improvement. Our code and datasets are publicly available (https://huggingface.co/datasets/sileod/mindgames , https://github.com/sileod/llm-theory-of-mind )