Evidence from counterfactual tasks supports emergent analogical reasoning in large language models
This defends the idea that LLMs have emergent reasoning skills, which is important for AI researchers but incremental as it responds to specific challenges.
The authors addressed critiques that large language models (LLMs) lack emergent analogical reasoning by showing they can generalize to counterfactual tasks with permuted alphabets, supporting their original claim of zero-shot capability.
We recently reported evidence that large language models are capable of solving a wide range of text-based analogy problems in a zero-shot manner, indicating the presence of an emergent capacity for analogical reasoning. Two recent commentaries have challenged these results, citing evidence from so-called `counterfactual' tasks in which the standard sequence of the alphabet is arbitrarily permuted so as to decrease similarity with materials that may have been present in the language model's training data. Here, we reply to these critiques, clarifying some misunderstandings about the test materials used in our original work, and presenting evidence that language models are also capable of generalizing to these new counterfactual task variants.