Reasoning models often fail on similar rule learning tasks
Thought without systematicity? Evaluating reasoning models on rule induction tasks
Computation and LanguageArtificial IntelligenceMachine Learning
Summary
The paper looks at whether reasoning models think in a way that lets them apply what they learn to slightly different but related problems. The authors tested these models on different versions of rule-learning tasks that are basically the same but with small changes. They found that even when models do well on one task, they often fail on similar tasks with the same underlying logic. This means these models might not understand or think as systematically as humans do.
What this means in practice
- •For ai system developers: Identify weaknesses in reasoning models by testing them on logically equivalent but varied rule learning tasks to improve their robustness.
- •For machine learning evaluators: Use the described task variations to more thoroughly assess model performance beyond single-context evaluations in rule-based reasoning systems.
Authors
Simon Schug, Brenden M. Lake
Abstract
A central tenet of human cognition is systematicity, the principle that understanding one concept is inherently tied to understanding close variations of that concept. Do reasoning models robustly exhibit such systematicity? If so, we would expect consistent performance on structurally equivalent variants of the same task. Here, we extend established rule induction tasks from cognitive science to assess the systematicity of thought in current reasoning models. Each task family has compositional structure that we use to create structurally equivalent task variations through task isomorphisms such as recombination and substitution. We find that despite being able to correctly solve a task, models often fail on structurally equivalent variants of the same task. These findings suggest that many model behaviors lack systematicity, rendering it difficult to robustly establish the cognitive abilities of reasoning models beyond the particular contexts they were evaluated in.