Ying, LanceCollins, Katherine M.Wei, MeganZhang, Cedegao E.Zhi-Xuan, TanWeller, AdrianTenenbaum, Joshua B.Wong, LionelCollins, Katherine2023-10-172023-06-27Ying, L., Collins, K. M., Wei, M., Zhang, C. E., Zhi-Xuan, T., Weller, A., ... & Wong, L. (2023). The Neuro-Symbolic Inverse Planning Engine (NIPE): Modeling Probabilistic Social Inferences from Linguistic Inputs. In Proceedings of the First Workshop on Theory of Mind in Communicating Agents, Honolulu, Hawaii, PMLR 202, 2023. arXiv preprint arXiv:2306.14325.https://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37377255Human beings are social creatures. We routinely reason about other agents, and a crucial component of this social reasoning is inferring people’s goals as we learn about their actions. In many settings, we can perform intuitive but reliable goal inference from language descriptions of agents, actions, and the background environments. In this paper, we study this process of language driving and influencing social reasoning in a probabilistic goal inference domain. We propose a neurosymbolic model that carries out goal inference from linguistic inputs of agent scenarios. The “neuro” part is a large language model (LLM) that translates language descriptions to code representations, and the “symbolic” part is a Bayesian inverse planning engine. To test our model, we design and run a human experiment on a linguistic goal inference task. Our model closely matches human response patterns and better predicts human judgements than using an LLM alone.en-USConference Paper2023-10-17Ying, L., Collins, K. M., Wei, M., Zhang, C. E., Zhi-Xuan, T., Weller, A., ... & Wong, L. (2023). The Neuro-Symbolic Inverse Planning Engine (NIPE): Modeling Probabilistic Social Inferences from Linguistic Inputs. arXiv preprint arXiv:2306.14325.