# Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control

arXiv: https://arxiv.org/abs/2512.16824 · HTML: https://mlsanity.com/paper/2512.16824/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 15 of its 24 references that are in the map.
- First posted on arXiv 18 Dec 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 15 of 15)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Learning an Approximate Model Predictive Controller with Guarantees](https://mlsanity.com/paper/1806.04167/) (2018)
- [Transformer-based Model Predictive Control: Trajectory Optimization via Sequence Modeling](https://mlsanity.com/paper/2410.23916/) (2024)
- [A Recursively Recurrent Neural Network (R2N2) Architecture for Learning Iterative Algorithms](https://mlsanity.com/paper/2211.12386/) (2022)
- [Improving Computational Efficiency for Powered Descent Guidance via Transformer-based Tight Constraint Prediction](https://mlsanity.com/paper/2311.05135/) (2023)
- [Fine-Tuned Language Models as Space Systems Controllers](https://mlsanity.com/paper/2501.16588/) (2025)
- [Multi-Phase Spacecraft Trajectory Optimization via Transformer-Based Reinforcement Learning](https://mlsanity.com/paper/2511.11402/) (2025)
- [Fine-tuning LLMs for Autonomous Spacecraft Control: A Case Study Using Kerbal Space Program](https://mlsanity.com/paper/2408.08676/) (2024)
- [Composing MPC with LQR and Neural Network for Amortized Efficiency and Stable Control](https://mlsanity.com/paper/2112.07238/) (2021)
- [A Hamilton-Jacobi Approach for Nonlinear Model Predictive Control in Applications with Navigational Uncertainty](https://mlsanity.com/paper/2503.23603/) (2025)
- [Action Chunking with Transformers for Image-Based Spacecraft Guidance and Control](https://mlsanity.com/paper/2509.04628/) (2025)

## Built on by (most cited first, 4 of 4)

- [Tiny Autoregressive Recursive Models](https://mlsanity.com/paper/2603.08082/) (2026)
- [Autonomous Reasoning for Spacecraft Control: A Large Language Model Framework with Group Relative Policy Optimization](https://mlsanity.com/paper/2601.04334/) (2026)
- [Tiny Recursive Models for Solving the J2-Perturbed Lambert Problem](https://mlsanity.com/paper/2606.00895/) (2026)
- [Agentic AutoResearch forSpace Autonomy: An Auditable, LLM-Driven Research Agent for Aerospace Control Problems](https://mlsanity.com/paper/2606.20394/) (2026)

## Abstract (arXiv)

Neural network controllers increasingly demand millions of parameters, and language model approaches push into the billions. For embedded aerospace systems with strict power and latency constraints, this scaling is prohibitive. We present Tiny Recursive Control (TRC), a neural architecture based on a counterintuitive principle: capacity can emerge from iteration depth rather than parameter count. TRC applies compact networks (approximately 1.5M parameters) repeatedly through a two-level hierarchical latent structure, refining control sequences by simulating trajectories and correcting based on tracking error. Because the same weights process every refinement step, adding iterations increases computation without increasing memory. We evaluate TRC on nonlinear control problems including oscillator stabilization and powered descent with fuel constraints. Across these domains, TRC achieves near-optimal control costs while requiring only millisecond-scale inference on GPU and under 10~MB memory, two orders of magnitude smaller than language model baselines. These results demonstrate that recursive reasoning, previously confined to discrete tasks, transfers effectively to continuous control synthesis.