Can Large Language Models generalize analogy solving like children can?
This highlights a key limitation in current LLMs for achieving human-like cognitive flexibility, which is incremental as it builds on prior work on analogy solving.
The study investigated whether large language models (LLMs) can generalize analogy solving to new domains like humans, finding that children and adults easily transferred knowledge to unfamiliar domains, but LLMs did not, indicating a lack of robust human-like analogical transfer.
In people, the ability to solve analogies such as "body : feet :: table : ?" emerges in childhood, and appears to transfer easily to other domains, such as the visual domain "( : ) :: < : ?". Recent research shows that large language models (LLMs) can solve various forms of analogies. However, can LLMs generalize analogy solving to new domains like people can? To investigate this, we had children, adults, and LLMs solve a series of letter-string analogies (e.g., a b : a c :: j k : ?) in the Latin alphabet, in a near transfer domain (Greek alphabet), and a far transfer domain (list of symbols). Children and adults easily generalized their knowledge to unfamiliar domains, whereas LLMs did not. This key difference between human and AI performance is evidence that these LLMs still struggle with robust human-like analogical transfer.