Boosting Theory-of-Mind Performance in Large Language Models via Prompting · HackerPapers discussions

Boosting Theory-of-Mind Performance in Large Language Models via Prompting — mentioned by famouswaffles

Tue, 11 Jul 2023 19:47:09 GMT

I don't know how we'd go about comparing reflection to human baselines but GPT-4 can reflect just fine https://arxiv.org/abs/2303.11366 GPT-4's emotional intelligence seems to be very high by the looks of things https://arxiv.org/abs/2304.11490 Empathy =/ Intelligence

Paper and discussion history. A mention is not an endorsement.

Boosting Theory-of-Mind Performance in Large Language Models via Prompting — mentioned by rodoxcasta

Fri, 12 May 2023 23:59:43 GMT

> There's no underlying theory of mind here. Actually, there's some experimental evidence that GPT4 have a Theory of Mind as good as humans, maybe better. https://arxiv.org/abs/2304.11490 > GPT-4 performed best in zero-shot settings, reaching nearly 80% ToM accuracy, but still fell short of the 87% human accuracy on the test set. However, when supplied with prompts for in-context learning, all RLHF-trained LLMs exceeded 80% ToM accuracy, with GPT-4 reaching 100%.

Paper and discussion history. A mention is not an endorsement.

Boosting Theory-of-Mind Performance in Large Language Models via Prompting — HN submission

Fri, 28 Apr 2023 19:01:15 GMT

Boosting Theory-of-Mind Performance in Large Language Models via Prompting

1 point; 3 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Boosting Theory-of-Mind Performance in Large Language Models via Prompting — HN submission

Tue, 25 Apr 2023 12:59:15 GMT

GPT4 can surpass humans in Theory of Mind test, with appropriate prompt

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.