XToM: Exploring the Multilingual Theory of Mind for Large Language Models
Chunkit Chan, Yauwai Yim, Hongchuan Zeng, Zhiying Zou, Xinyuan Cheng, Zhifan Sun, Zheye Deng, Kawai Chung, Yuzhuo Ao, Yixiang Fan, Cheng Jiayang, Ercong Nie
摘要
Theory of Mind (ToM)-the ability to infer mental states in others-is pivotal for human social cognition. Existing evaluations of ToM in LLMs are largely limited to English, neglecting the linguistic diversity that shapes human cognition. This limitation raises a critical question: can LLMs exhibit Multilingual Theory of Mind-the capacity to reason about mental states across diverse linguistic contexts? To address this gap, we present XToM, a rigorously validated multilingual benchmark that evaluates ToM across five languages and incorporates diverse, contextually rich task scenarios. Using XToM, we systematically evaluate LLMs (e.g., DeepSeek R1), revealing a pronounced dissonance: while models excel in multilingual language understanding, their ToM performance varies across languages. Our findings expose limitations in LLMs' ability to replicate human-like mentalizing across linguistic contexts 1 . * Equal contribution. 1 Code and data are available at https://github.com/ HKUST-KnowComp/XToM . …… … XToMi 1 Personne 1 est entrée dans la chambre principale. 2 Personne 2 est entrée dans la chambre principale. 3 L'oignon est dans la bouteille. 4 Personne 2 est sortie de la chambre principale. 5 Personne 1 a déplacé l'oignon dans le garde-manger. Où Personne 2 cherchera-t-elle l'oignon ? A) Garde-manger B) Bouteille FR XToMi 1 人物1进入主卧室。 2 人物2进入主卧室。 3 洋葱在瓶子里。 4 人物2离开主卧室。 5 人物1将洋葱移到储藏室。 人物2会去哪里找洋葱? A) 储藏室 B) 瓶子 ZH XToMi -BeliefQ 1 Person 1 entered the master bedroom. 2 Person 2 entered the master bedroom. 3 The onion is in the bottle. 4 Person 2 exited the master bedroom. 5 Person 1 moved the onion to the pantry. Where will Person 2 look for the onion? A) Pantry B) Bottle EN …… … XToMi 1 Personne 1 est entrée dans la chambre principale. 2 Personne 2 est entrée dans la chambre principale. 3 L'oignon est dans la bouteille. 4 Personne 2 est sortie de la chambre principale. 5 Personne 1 a déplacé l'oignon dans le garde-manger. Où Personne 2 cherchera-t-elle l'oignon ? A) Garde-manger B) Bouteille FR XToMi 1 人物1进入主卧室。 2 人物2进入主卧室。 3 洋葱在瓶子里。 4 人物2离开主卧室。 5 人物1将洋葱移到储藏室。 人物2会去哪里找洋葱?? A) 储藏室 B) 瓶子 ZH EN XToMi -RealityQ 1 Person 1 entered the master bedroom. 2 Person 2 entered the master bedroom. 3 The onion is in the bottle. 4 Person 2 exited the master bedroom. 5 Person 1 moved the onion to the pantry.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
它引用的顶会 Paper27
- Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida 等NeurIPS 2022 · 被引用 24,707 次
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma 等NeurIPS 2022 · 被引用 22,562 次
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu 等ICLR 2022 · 被引用 18,833 次
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger 等ICLR 2020 · 被引用 8,443 次
- Climbing towards NLU: On Meaning, Form, and Understanding in the Age of DataEmily M. Bender, Alexander KollerACL 2020 · 被引用 914 次
相关 Paper
- ToMBench: Benchmarking Theory of Mind in Large Language ModelsZhuang Chen, Jincenzi Wu, Jinfeng Zhou, Bosi Wen 等ACL 2024 · 被引用 6 次
- The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story CharactersChulun Zhou, Qiujing Wang, Mo Yu, Xiaoqian Yue 等ACL 2025 · 被引用 9 次
- Perceptions to Beliefs: Exploring Precursory Inferences for Theory of Mind in Large Language ModelsChani Jung, Dongkwan Kim, Jiho Jin, Jiseon Kim 等EMNLP 2024 · 被引用 2 次
- Towards Dynamic Theory of Mind: Evaluating LLM Adaptation to Temporal Evolution of Human StatesYang Xiao, Jiashuo Wang, Qiancheng Xu, Changhe Song 等ACL 2025 · 被引用 12 次
- Theory of Mind in Large Language Models: Assessment and EnhancementRuirui Chen, Weifeng Jiang, Chengwei Qin, Cheston TanACL 2025
