ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915104606388224 |
|---|---|
| author | Shinoda, Kazutoshi Hojo, Nobukatsu Nishida, Kyosuke Mizuno, Saki Suzuki, Keita Masumura, Ryo Sugiyama, Hiroaki Saito, Kuniko |
| author_facet | Shinoda, Kazutoshi Hojo, Nobukatsu Nishida, Kyosuke Mizuno, Saki Suzuki, Keita Masumura, Ryo Sugiyama, Hiroaki Saito, Kuniko |
| contents | Existing Theory of Mind (ToM) benchmarks diverge from real-world scenarios in three aspects: 1) they assess a limited range of mental states such as beliefs, 2) false beliefs are not comprehensively explored, and 3) the diverse personality traits of characters are overlooked. To address these challenges, we introduce ToMATO, a new ToM benchmark formulated as multiple-choice QA over conversations. ToMATO is generated via LLM-LLM conversations featuring information asymmetry. By employing a prompting method that requires role-playing LLMs to verbalize their thoughts before each utterance, we capture both first- and second-order mental states across five categories: belief, intention, desire, emotion, and knowledge. These verbalized thoughts serve as answers to questions designed to assess the mental states of characters within conversations. Furthermore, the information asymmetry introduced by hiding thoughts from others induces the generation of false beliefs about various mental states. Assigning distinct personality traits to LLMs further diversifies both utterances and thoughts. ToMATO consists of 5.4k questions, 753 conversations, and 15 personality trait patterns. Our analysis shows that this dataset construction approach frequently generates false beliefs due to the information asymmetry between role-playing LLMs, and effectively reflects diverse personalities. We evaluate nine LLMs on ToMATO and find that even GPT-4o mini lags behind human performance, especially in understanding false beliefs, and lacks robustness to various personality traits. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_08838 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind Shinoda, Kazutoshi Hojo, Nobukatsu Nishida, Kyosuke Mizuno, Saki Suzuki, Keita Masumura, Ryo Sugiyama, Hiroaki Saito, Kuniko Computation and Language Artificial Intelligence Existing Theory of Mind (ToM) benchmarks diverge from real-world scenarios in three aspects: 1) they assess a limited range of mental states such as beliefs, 2) false beliefs are not comprehensively explored, and 3) the diverse personality traits of characters are overlooked. To address these challenges, we introduce ToMATO, a new ToM benchmark formulated as multiple-choice QA over conversations. ToMATO is generated via LLM-LLM conversations featuring information asymmetry. By employing a prompting method that requires role-playing LLMs to verbalize their thoughts before each utterance, we capture both first- and second-order mental states across five categories: belief, intention, desire, emotion, and knowledge. These verbalized thoughts serve as answers to questions designed to assess the mental states of characters within conversations. Furthermore, the information asymmetry introduced by hiding thoughts from others induces the generation of false beliefs about various mental states. Assigning distinct personality traits to LLMs further diversifies both utterances and thoughts. ToMATO consists of 5.4k questions, 753 conversations, and 15 personality trait patterns. Our analysis shows that this dataset construction approach frequently generates false beliefs due to the information asymmetry between role-playing LLMs, and effectively reflects diverse personalities. We evaluate nine LLMs on ToMATO and find that even GPT-4o mini lags behind human performance, especially in understanding false beliefs, and lacks robustness to various personality traits. |
| title | ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2501.08838 |