@taylorwwebb : Have LLMs developed an emergent reasoning capacity, or are they merely mimicking their training data? In this new preprint, we reply to recent commentaries that employed counterfactual analogy tasks, finding that GPT-4 *can* generalize to these new tasks. arxiv.org/abs/24504.13070 • TwiCopy

Taylor Webb

@taylorwwebb

+ Follow

Studying cognition in humans and machines.

ID: 921368137668362240

linkhttps://scholar.google.com/citations?user=WCmrJoQAAAAJ&hl=en calendar_today20-10-2017 13:30:57

393 Tweet

736 Followers

512 Following

Taylor Webb

@taylorwwebb

6 months ago

Have LLMs developed an emergent reasoning capacity, or are they merely mimicking their training data? In this new preprint, we reply to recent commentaries that employed counterfactual analogy tasks, finding that GPT-4 *can* generalize to these new tasks. arxiv.org/abs/2404.13070

thumb_up_off_alt41

chat_bubble_outline2

repeat6

shareShare