# Prospective Compression in Human Abstraction Learning

arXiv: https://arxiv.org/abs/2605.09985 · HTML: https://mlsanity.com/paper/2605.09985/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 41 references that are in the map.
- First posted on arXiv 11 May 2026.

## Builds on (most cited first, 17 of 17)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning](https://mlsanity.com/paper/2006.08381/) (2020)
- [Hypothesis Search: Inductive Reasoning with Language Models](https://mlsanity.com/paper/2309.05660/) (2023)
- [Top-Down Synthesis for Library Learning](https://mlsanity.com/paper/2211.16605/) (2022)
- [LILO: Learning Interpretable Libraries by Compressing and Documenting Code](https://mlsanity.com/paper/2310.19791/) (2023)
- [Communicating Natural Programs to Humans and Machines](https://mlsanity.com/paper/2106.07824/) (2021)
- [Program Synthesis and Semantic Parsing with Learned Code Idioms](https://mlsanity.com/paper/1906.10816/) (2019)
- [Fast and flexible: Human program induction in abstract reasoning tasks](https://mlsanity.com/paper/2103.05823/) (2021)
- [Learning abstract structure for drawing by efficient motor program induction](https://mlsanity.com/paper/2008.03519/) (2020)
- [babble: Learning Better Abstractions with E-Graphs and Anti-Unification](https://mlsanity.com/paper/2212.04596/) (2022)
- [Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines](https://mlsanity.com/paper/2205.11558/) (2022)
- [Exploring the hierarchical structure of human plans via program generation](https://mlsanity.com/paper/2311.18644/) (2023)
- [Map Induction: Compositional spatial submap learning for efficient exploration in novel environments](https://mlsanity.com/paper/2110.12301/) (2021)
- [Cognitive maps are generative programs](https://mlsanity.com/paper/2504.20628/) (2025)
- [Online library learning in human visual puzzle solving](https://mlsanity.com/paper/2603.23244/) (2026)
- [Library learning with e-graphs on jazz harmony](https://mlsanity.com/paper/2605.04622/) (2026)

## Abstract (arXiv)

A core challenge in program synthesis is online library learning: the incremental acquisition of reusable abstractions under uncertainty about future task demands. Existing algorithms treat library learning as retrospective compression over a static task distribution, where the learned library is determined by the corpus of past tasks. However, real-world learning domains are often non-stationary, with tasks arising from a generative process that evolves over time. We propose and test the hypothesis that in non-stationary domains human library learning selects abstractions prospectively: targeting compression of future tasks. We study this question using the Pattern Builder Task, a visual program synthesis paradigm in which participants construct increasingly complex geometric patterns from a small set of primitives, transformations, and custom helpers that carry forward across trials. Using this task, we conduct two experiments with complementary latent curricula, designed to dissociate between behaviors consistent with prospective compression, and alternative library learning accounts. Using six computational models spanning online library learning strategies, we show that human abstraction behavior reflects sensitivity to latent, non-stationary structure in the task-generating process. This behavior is consistent with prospective compression, and cannot be captured by existing retrospective compression-based algorithms, or inductive biases modeled by LLM-based program synthesis.