# Before Parc Ferm'e: RL-Time Pruning for Efficient Embodied LLMs in Autonomous Driving

arXiv: https://arxiv.org/abs/2605.31256 · HTML: https://mlsanity.com/paper/2605.31256/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 23 of its 25 references that are in the map.
- First posted on arXiv 29 May 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- End-to-end autonomous driving / [Large language models for driving](https://mlsanity.com/driving/large-language-models-for-driving/)

## Builds on (most cited first, 23 of 23)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Code as Policies: Language Model Programs for Embodied Control](https://mlsanity.com/paper/2209.07753/) (2022)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [Inner Monologue: Embodied Reasoning through Planning with Language Models](https://mlsanity.com/paper/2207.05608/) (2022)
- [A Simple and Effective Pruning Approach for Large Language Models](https://mlsanity.com/paper/2306.11695/) (2023)
- [LLM-Pruner: On the Structural Pruning of Large Language Models](https://mlsanity.com/paper/2305.11627/) (2023)
- [DriveLM: Driving with Graph Visual Question Answering](https://mlsanity.com/paper/2312.14150/) (2023)
- [LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models](https://mlsanity.com/paper/2212.04088/) (2022)
- [DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Model](https://mlsanity.com/paper/2310.01412/) (2023)
- [Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning](https://mlsanity.com/paper/2310.06694/) (2023)
- [SliceGPT: Compress Large Language Models by Deleting Rows and Columns](https://mlsanity.com/paper/2401.15024/) (2024)
- [Fluctuation-based Adaptive Structured Pruning for Large Language Models](https://mlsanity.com/paper/2312.11983/) (2023)
- [Drive as You Speak: Enabling Human-Like Interaction with Large Language Models in Autonomous Vehicles](https://mlsanity.com/paper/2309.10228/) (2023)
- [A Survey on Efficient Vision-Language-Action Models](https://mlsanity.com/paper/2510.24795/) (2025)
- [CFSP: An Efficient Structured Pruning Framework for LLMs with Coarse-to-Fine Activation Information](https://mlsanity.com/paper/2409.13199/) (2024)
- [Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models](https://mlsanity.com/paper/2504.11514/) (2025)
- [RL-Pruner: Structured Pruning Using Reinforcement Learning for CNN Compression and Acceleration](https://mlsanity.com/paper/2411.06463/) (2024)
- [ForzaETH Race Stack -- Scaled Autonomous Head-to-Head Racing on Fully Commercial off-the-Shelf Hardware](https://mlsanity.com/paper/2403.11784/) (2024)
- [Adapt-Pruner: Adaptive Structural Pruning for Efficient Small Language Model Training](https://mlsanity.com/paper/2502.03460/) (2025)
- [NARRATE: Versatile Language Architecture for Optimal Control in Robotics](https://mlsanity.com/paper/2403.10762/) (2024)
- [RobotxR1: Enabling Embodied Robotic Intelligence on Large Language Models through Closed-Loop Reinforcement Learning](https://mlsanity.com/paper/2505.03238/) (2025)

## Built on by (most cited first, 2 of 2)

- [Post-Training in End-to-End Autonomous Driving](https://mlsanity.com/paper/2607.08072/) (2026)
- [TAP: Efficient Long-Horizon Agent Pruning via Trajectory-Anchored Recovery](https://mlsanity.com/paper/2610.09074/) (2026)

## Abstract (arXiv)

Embodied Large Language Models (LLMs) are increasingly used as reasoning modules in robotic control pipelines to improve human-robot interaction, but their memory and generation latency make real-time deployment difficult. Pruning can reduce these costs, but for controllers that undergo multiple pre- and post-training phases, the crucial question is not only how much to prune, but when pruning should occur. In this work, we propose Before Parc Ferm'e (BPF), a pruning strategy performed during RL that compresses embodied LLM controllers while they are still being optimized for closed-loop behavior. This allows pruning decisions to account for the task-specific supervision and closed-loop feedback that shape the final controller. We propose two variants: BPF-RL, which performs iterative pruning during RL by removing part of the model at predefined training intervals, and BPF-SFT/RL, which first prunes part of the model structure during SFT and then further compresses it during RL using the same iterative strategy as BPF-RL until the target pruning ratio is reached. We evaluate BPF on RobotxR1, an LLM-based autonomous-driving control pipeline, using an established LLM pruning framework (LLM-Pruner), and compare it against post-training pruning, post-training pruning with RL recovery, SFT-stage pruning, and smaller dense models from the same family. Our results show that BPF provides the best task-performance vs. memory and throughput trade-off among the considered pruning strategies. When compressing the larger RobotxR1 models, BPF-SFT/RL achieves a 1.69× better size-end-to-end performance trade-off than directly selecting a smaller dense model from the same family, measured as removed parameters per lost percentage point of control adaptability. On the Jetson AGX Orin mounted on the target robotic platform, the compact models improve decode throughput by up to 27%.