# Searching Latent Program Spaces

arXiv: https://arxiv.org/abs/2411.08706 · HTML: https://mlsanity.com/paper/2411.08706/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Nov 2024.

## Built on by (most cited first, 20 of 20)

- [ARC Prize 2024: Technical Report](https://mlsanity.com/paper/2412.04604/) (2024)
- [Neural Cellular Automata for ARC-AGI](https://mlsanity.com/paper/2506.15746/) (2025)
- [ARC-GEN: A Mimetic Procedural Benchmark Generator for the Abstraction and Reasoning Corpus](https://mlsanity.com/paper/2511.00162/) (2025)
- [Think Visually, Reason Textually: Vision-Language Synergy in ARC](https://mlsanity.com/paper/2511.15703/) (2025)
- [Test-time Adaptation of Tiny Recursive Models](https://mlsanity.com/paper/2511.02886/) (2025)
- [Gradient-Based Program Synthesis with Neurally Interpreted Languages](https://mlsanity.com/paper/2604.18907/) (2026)
- [Gradient-Based Program Repair: Fixing Bugs in Continuous Program Spaces](https://mlsanity.com/paper/2505.17703/) (2025)
- [Context Tuning for In-Context Optimization](https://mlsanity.com/paper/2507.04221/) (2025)
- [ARC-AGI Without Pretraining](https://mlsanity.com/paper/2512.06104/) (2025)
- [Disentangling Dynamical Systems: Causal Representation Learning Meets Local Sparse Attention](https://mlsanity.com/paper/2603.14483/) (2026)
- [Learning to Disentangle Latent Reasoning Rules with Language VAEs: A Systematic Study](https://mlsanity.com/paper/2506.19418/) (2025)
- [Bridging Reasoning to Learning: Unmasking Illusions using Complexity Out of Distribution Generalization](https://mlsanity.com/paper/2510.06274/) (2025)
- [Latent Programming Horizons in Coding Agents](https://mlsanity.com/paper/2607.05188/) (2026)
- [Improving Generative Inverse Design of Rectangular Patch Antennas with Test Time Optimization](https://mlsanity.com/paper/2505.18188/) (2025)
- [Decoupling Search and Learning in Neural Net Training](https://mlsanity.com/paper/2509.10973/) (2025)
- [Latent Heuristic Search: Continuous Optimization for Automated Algorithm Design](https://mlsanity.com/paper/2605.17137/) (2026)
- [FILLER: Feature Imputation via Latent Location Exploration and Retrieval](https://mlsanity.com/paper/2607.23295/) (2026)
- [Position: Reasoning is a Learnable Rule-Based Process](https://mlsanity.com/paper/2608.12325/) (2026)
- [Intuition vectors](https://mlsanity.com/paper/2609.32619/) (2026)
- [On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D](https://mlsanity.com/paper/2610.00820/) (2026)

## Abstract (arXiv)

General intelligence requires systems that acquire new skills efficiently and generalize beyond their training distributions. Although program synthesis approaches have strong generalization power, they face scaling issues due to the large combinatorial spaces that quickly render them impractical, requiring human-generated DSLs or pre-trained priors to narrow this search space. On the other hand, deep learning methods have had high successes, but they lack structured test-time adaptation and rely on heavy stochastic sampling or expensive gradient updates for fine-tuning. In this work, we propose the Latent Program Network (LPN), a novel architecture that builds in test-time search directly into neural models. LPN learns a latent space of implicit programs -- neurally mapping inputs to outputs -- through which it can search using gradients at test time. LPN combines the adaptability of symbolic approaches and the scalability of neural methods. It searches through a compact latent space at test time and bypasses the need for pre-defined domain-specific languages. On a range of programming-by-examples tasks, LPN either outperforms or matches performance compared to in-context learning and test-time training methods. Tested on the ARC-AGI benchmark, we demonstrate that LPN can both learn a compact program space and search through it at test time to adapt to novel tasks. LPN doubles its performance on out-of-distribution tasks when test-time search is switched on.