mlsanityData to 7 Oct 2026

RLRC: Reinforcement Learning-based Recovery for Compressed Vision-Language-Action Models

arXiv 2506.17639 ↗First posted 21 Jun 2025cs.RO cs.AIcited by 20

On the map: Vision-language-action models → Reinforcement learning and post-training

Abstract (arXiv)

Vision-Language-Action models (VLA) have demonstrated remarkable capabilities and strong potential in complex robotic manipulation. However, their large parameter sizes and high inference latency hinder real-world deployment, especially on resource-constrained platforms. To address this, we conduct a systematic empirical study of model compression for VLAs. Building on these insights, we present textitRLRC, a three-stage compression and recovery pipeline consisting of structured pruning, performance recovery via SFT and RL, and subsequent quantization. The RL stage incorporates a critic warm-up strategy and BC loss regularization to stabilize training and preserve policy behavior. RLRC achieves up to an 8 times memory reduction and 2.3 times inference speedup while maintaining the original task success rate. Extensive experiments across multiple VLA backbones show that RLRC consistently outperforms existing compression baselines, highlighting its effectiveness for on-device deployment. Project website: https://rlrc-vla.github.io

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Placed in Reinforcement learning and post-training, one of 13 directions in Vision-language-action models. Builds on 36 of its 41 references that are in the map. First posted on arXiv 21 Jun 2025.

Builds on

36 of 41 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025What Can RL Bring to VLA Generalization? An Empirical Study26 May
  2. 2025VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning24 May
  3. 2025Interactive Post-Training for Vision-Language-Action Models22 May
  4. 2025Improving Vision-Language-Action Model with Online Reinforcement Learning28 Jan
  5. 2024Octo: An Open-Source Generalist Robot Policy20 May

Other references · 31

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,355
  3. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  4. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  5. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  6. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,838
  7. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
  8. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  9. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  10. 2022LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scalecited by 881
  11. 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 849
  12. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
19 more
  1. 2023A Simple and Effective Pruning Approach for Large Language Modelscited by 776
  2. 2023LLM-Pruner: On the Structural Pruning of Large Language Modelscited by 643
  3. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  4. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Roboticscited by 507
  5. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
  6. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  7. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 331
  8. 2023LLM-QAT: Data-Free Quantization Aware Training for Large Language Modelscited by 320
  9. 2024ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AIcited by 216
  10. 2024RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulationcited by 166
  11. 2023Fluctuation-based Adaptive Structured Pruning for Large Language Modelscited by 110
  12. 2025EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Modelscited by 89
  13. 2025VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Cachingcited by 83
  14. 2025Universal Actions for Enhanced Embodied Foundation Modelscited by 80
  15. 2025SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Accelerationcited by 56
  16. 2022In-context Learning Distillation: Transferring Few-shot Learning Ability of Pre-trained Language Modelscited by 36
  17. 2025Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Modelscited by 30
  18. 2024Quantization-Aware Imitation-Learning for Resource-Efficient Robotic Controlcited by 25
  19. 2024FASTNav: Fine-tuned Adaptive Small-language-models Trained for Multi-point Robot Navigationcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 14.

20 papers.

Sort:

By area:

  1. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applicationscited by 82
  2. 2025VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulatorscited by 65
  3. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46
  4. 2025A Survey on Efficient Vision-Language-Action Modelscited by 34
  5. 2026FASTER: Rethinking Real-Time Flow VLAscited by 20
  6. 2025DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Modelscited by 17
  7. 2026TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAMcited by 9
  8. 2026DySL-VLA: Efficient Vision-Language-Action Model Inference via Dynamic-Static Layer-Skipping for Robot Manipulationcited by 8
  9. 2026StreamingVLA: Streaming Vision-Language-Action Model with Action Flow Matching and Adaptive Early Observationcited by 6
  10. 2026Vesta: A Generalist Embodied Reasoning Modelcited by 5
  11. 2026SAFE-Pruner: Semantic Attention-Guided Future-Aware Token Pruning for Efficient Vision-Language-Action Manipulationcited by 4
  12. 2025Token Compression Meets Compact Vision Transformers: A Survey and Comparative Evaluation for Edge AIcited by 3
8 more
  1. 2026Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?cited by 3
  2. 2026Revisiting Parameter Redundancy in Vision-Language-Action Models: Insights from VLM-to-VLA Adaptationcited by 2
  3. 2026Environment-Aware Adaptive Pruning with Interleaved Inference Orchestration for Vision-Language-Action Modelscited by 1
  4. 2026Joint Architecture-Token-Bitwidth Multi-Axis Optimization of Vision Transformers for Semiconductor IC Packagingcited by 0
  5. 2026IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policiescited by 0
  6. 2026Recovering Aggressively Pruned Vision-Language-Action Models with Offline Hidden-State Distillationcited by 0
  7. 2026Anatomy of a Closed-Loop Collapse: A Causal Case Study of a Compressed VLA Policycited by 0
  8. 2026REACT: Rolling Denoising and Dual Decoupling for Reactive Robot Control with VLA Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Finetuning Vision-Language-Action Models Requires Fewer Layers Than You Think
  2. 2026Reducing Temporal Redundancy for Efficient Vision-Language-Action Inference
  3. 2026VLaRL: Augmenting Vision-Language-Action Models with Simulation-Trained Latent-Conditioned Residual RL
  4. 2025SRPO: Self-Referential Policy Optimization for Vision-Language-Action Models
  5. 2026TEMPO: Semantic-Action Decoupled RL Post-Training for Vision-Language-Action Models
  6. 2026Z-1: Efficient Reinforcement Learning for Vision-Language-Action Models
  7. 2026Learned Image Compression for Vision-Language-Action Models
  8. 2026Learning to Accelerate Vision-Language-Action Models through Adaptive Visual Token Caching