Can Large Language Models generalize analogy solving like children can? · HackerPapers discussions

Can Large Language Models generalize analogy solving like children can? — mentioned by gus_massa

Thu, 18 Dec 2025 02:07:14 GMT

Preprint: https://arxiv.org/pdf/2411.02348 > For children, the interface was explained and demonstrated in person. Participants then solved two simple practice items without feedback (used to ensure understanding of task). I'm still not sure if the prompt for children is equivalent to the prompt for AI. Anyway, it's hard for the current AI to do these tasks, like counting the r in "strawberry" because tokenization hides a lot of details. Anyway, AI is improving and next year AI will under the hood cut and paste the list of characters in a Python script that solves the task.

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

Can Large Language Models generalize analogy solving like children can? — HN submission

Wed, 17 Dec 2025 12:19:55 GMT

Can large language models generalize analogy solving like children can? [pdf]

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

Can Large Language Models generalize analogy solving like children can? — HN submission

Tue, 05 Nov 2024 15:02:55 GMT

Can Large Language Models generalize analogy solving like people can?

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