mlsanityData to 7 Oct 2026

SQAP-VLA: A Synergistic Quantization-Aware Pruning Framework for High-Performance Vision-Language-Action Models

arXiv 2509.09090 ↗First posted 11 Sep 2025cs.CV cs.AIcited by 15

On the map: Vision-language-action models → Efficient VLA inference and adaptation

also in 1 more
Abstract (arXiv)

Vision-Language-Action (VLA) models exhibit unprecedented capabilities for embodied intelligence. However, their extensive computational and memory costs hinder their practical deployment. Existing VLA compression and acceleration approaches conduct quantization or token pruning in an ad-hoc manner but fail to enable both for a holistic efficiency improvement due to an observed incompatibility. This work introduces SQAP-VLA, the first structured, training-free VLA inference acceleration framework that simultaneously enables state-of-the-art quantization and token pruning. We overcome the incompatibility by co-designing the quantization and token pruning pipeline, where we propose new quantization-aware token pruning criteria that work on an aggressively quantized model while improving the quantizer design to enhance pruning effectiveness. When applied to standard VLA models, SQAP-VLA yields significant gains in computational efficiency and inference speed while successfully preserving core model performance, achieving a ×1.93 speedup and up to a 4.5% average success rate enhancement compared to the original model.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Placed in Efficient VLA inference and adaptation, one of 13 directions in Vision-language-action models. Builds on 31 of its 35 references that are in the map. First posted on arXiv 11 Sep 2025.

Builds on

31 of 35 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration15 Jun
  2. 2025EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models11 Jun
  3. 2025MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation26 Mar
  4. 2025VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching4 Feb
  5. 2024Quantization-Aware Imitation-Learning for Resource-Efficient Robotic Control2 Dec

Other references · 26

  1. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  2. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  3. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  4. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  5. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,838
  6. 2022Token Merging: Your ViT But Fastercited by 995
  7. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  8. 2024An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Modelscited by 629
  9. 2021A White Paper on Neural Network Quantizationcited by 453
  10. 2024SpinQuant: LLM quantization with learned rotationscited by 446
  11. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  12. 2023OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Modelscited by 403
14 more
  1. 2024QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebookscited by 354
  2. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  3. 2023A Survey on Model Compression for Large Language Modelscited by 263
  4. 2021Pruning and Quantization for Deep Neural Network Acceleration: A Surveycited by 259
  5. 2020SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruningcited by 228
  6. 2024QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Servingcited by 200
  7. 2024SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Modelscited by 161
  8. 2024Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMscited by 153
  9. 2024Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Modelscited by 125
  10. 2023Revisiting Token Pruning for Object Detection and Instance Segmentationcited by 50
  11. 2023Pruning vs Quantization: Which is Better?cited by 45
  12. 2021Ps and Qs: Quantization-aware pruning for efficient low latency neural network inferencecited by 31
  13. 2020Single Shot 6D Object Pose Estimationcited by 19
  14. 2025FBQuant: FeedBack Quantization for Large Language Modelscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 12.

15 papers.

Sort:

By area:

  1. 2025A Survey on Efficient Vision-Language-Action Modelscited by 34
  2. 2025Efficient Vision-Language-Action Models for Embodied Manipulation: A Systematic Surveycited by 25
  3. 2026FASTER: Rethinking Real-Time Flow VLAscited by 20
  4. 2026DyQ-VLA: Temporal-Dynamic-Aware Quantization for Embodied Vision-Language-Action Modelscited by 10
  5. 2026A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Modelcited by 8
  6. 2026AC^2-VLA: Action-Context-Aware Adaptive Computation in Vision-Language-Action Models for Efficient Robotic Manipulationcited by 6
  7. 2025An Anatomy of Vision-Language-Action Models: From Modules to Milestones and Challengescited by 4
  8. 2026OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelismcited by 4
  9. 2026From Inference Efficiency to Embodied Efficiency: Revisiting Efficiency Metrics for Vision-Language-Action Modelscited by 2
  10. 2026Revisiting Parameter Redundancy in Vision-Language-Action Models: Insights from VLM-to-VLA Adaptationcited by 2
  11. 2026Beyond Attention Magnitude: Leveraging Inter-layer Rank Consistency for Efficient Vision-Language-Action Modelscited by 1
  12. 2026Towards Joint Quantization and Token Pruning of Vision-Language Modelscited by 1
3 more
  1. 2026Just Noticeable Difference Modeling for Token Compression in Vision-Language-Action Modelscited by 0
  2. 2026VLAQuantBench: Closed-Loop Evaluation of Post-Training Quantization for Vision-Language-Action Modelscited by 0
  3. 2026P4Q: Co-designing Token Pruning and Quantization for Vision-Language Model Accelerationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Bridging the Semantic-Action Gap in Visual Token Pruning for Efficient VLA Inference
  2. 2025SpecPrune-VLA: Accelerating Vision-Language-Action Models via Action-Aware Self-Speculative Pruning
  3. 2025The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruning
  4. 2025Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models
  5. 2025Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models
  6. 2026QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models
  7. 2025EaqVLA: Encoding-aligned Quantization for Vision-Language-Action Models
  8. 2026QVLA: Not All Channels Are Equal in Vision-Language-Action Model's Quantization