Meta-Learning Metaprogram Primitives for Program Induction on List Manipulation Tasks

May 12, 2026 • Yiding Song, Christopher Bates, Kazuki Irie, Samuel Gershman (Final Report, CS 91r)

Abstract

Program induction represents an important skill in abstracting away from observations to explaining them with programs, believed to enable faster concept-learning and generalisation in humans. Traditionally, efforts at eliciting such capabilities in machines have been achieved through a variety of neuro-symbolic methods. However, given the success of meta-learning frameworks on systematic generalisation (Lake and Baroni, 2023), we propose to study program induction through a meta-learning framework instead. Our preliminary results demonstrate that Transformers can learn to perform program induction on list-manipulation tasks, with meta-learning boosting accuracy in data-limited settings. Meanwhile, regardless of dataset size, meta-learned models tend to form representations that align more closely to metaprogram primitives ('metaprimitives') than their in-weight counterparts. This suggests that meta-learning encourages Transformers to develop qualitatively different algorithms to program induction problems, potentially favouring metaprimitive-related solutions that have been shown to be effective symbolic models of human behaviour (Rule et al., 2024).