Using Counterfactual Tasks to Evaluate the Generality of Analogical Reasoning in Large Language Models
This work addresses the problem of evaluating the generality of reasoning in AI models for researchers and developers, showing incremental evidence against robust analogy-making in LLMs.
The study investigated whether large language models (LLMs) perform humanlike abstract analogical reasoning by testing them on original and counterfactual analogy problems, finding that while humans maintained high performance, GPT models' performance declined sharply on counterfactual tasks.
Large language models (LLMs) have performed well on several reasoning benchmarks, including ones that test analogical reasoning abilities. However, it has been debated whether they are actually performing humanlike abstract reasoning or instead employing less general processes that rely on similarity to what has been seen in their training data. Here we investigate the generality of analogy-making abilities previously claimed for LLMs (Webb, Holyoak, & Lu, 2023). We take one set of analogy problems used to evaluate LLMs and create a set of "counterfactual" variants-versions that test the same abstract reasoning abilities but that are likely dissimilar from any pre-training data. We test humans and three GPT models on both the original and counterfactual problems, and show that, while the performance of humans remains high for all the problems, the GPT models' performance declines sharply on the counterfactual set. This work provides evidence that, despite previously reported successes of LLMs on analogical reasoning, these models lack the robustness and generality of human analogy-making.