2000 character limit reached
Spontaneous Theory of Mind for Artificial Intelligence (2402.13272v1)
Published 16 Feb 2024 in cs.AI and cs.HC
Abstract: Existing approaches to Theory of Mind (ToM) in AI overemphasize prompted, or cue-based, ToM, which may limit our collective ability to develop Artificial Social Intelligence (ASI). Drawing from research in computer science, cognitive science, and related disciplines, we contrast prompted ToM with what we call spontaneous ToM -- reasoning about others' mental states that is grounded in unintentional, possibly uncontrollable cognitive functions. We argue for a principled approach to studying and developing AI ToM and suggest that a robust, or general, ASI will respond to prompts \textit{and} spontaneously engage in social reasoning.
- Baron-Cohen, S., Leslie, A.M., Frith, U.: Does the autistic child have a “theory of mind”? Cognition 21(1), 37–46 (1985)
- Ericsson, K.A.: Protocol analysis. A companion to cognitive science pp. 425–432 (2017)
- Fodor, J.A.: The modularity of mind. MIT press (1983)
- Frith, U., Happé, F.: Autism: Beyond “theory of mind”. Cognition 50(1-3), 115–132 (1994)
- Gerson, L.P.: Knowing Persons: a study in Plato. Clarendon Press (2003)
- Kosinski, M.: Theory of mind may have spontaneously emerged in large language models. arXiv preprint arXiv:2302.02083 (2023)
- Rozemond, M.: Descartes’s dualism. Harvard University Press (2009)
- Ullman, T.: Large language models fail on trivial alterations to theory-of-mind tasks. arXiv preprint arXiv:2302.08399 (2023)
- Nikolos Gurney (7 papers)
- David V. Pynadath (7 papers)
- Volkan Ustun (5 papers)