Cameron Jones
I'm interested in social intelligence in humans and LLMs. My recent work looks at LLM performance on the False Belief task, other theory of mind tasks, and the Turing test.
I'm interested in social intelligence in humans and LLMs. My recent work looks at LLM performance on the False Belief task, other theory of mind tasks, and the Turing test.
Jones, C. R. & Bergen, B. (under review). People cannot distinguish GPT-4 from a human in a Turing test. [preprint]
Jones, C. R. & Bergen, B. (to appear). Does GPT-4 pass the Turing test?. NAACL 2024. [preprint]
Jones, C. R., Trott, S., & Bergen, B. (to appear). Comparing Humans and Large Language Models on an Experimental Protocol Inventory for Theory of Mind Evaluation (EPITOME). Transactions of the Association of Computational Linguistics. [preprint]
Trott, S.*, Jones, C. R.*, Chang, T., Michaelov, M., & Bergen, B. (2023). Do Large Language Models know what humans know?. Cognitive Science, 47(7). [*equal contribution] [paper]
Understanding, Grounding, and Reference in LLMs, with Sean Trott. The Gradient Podcast
Does GPT-3 Have a Theory of Mind? Cognitive Science Society Blog