Research map / Vision-language and multimodal / Robot action and navigation
SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration
Ye Li, Yuan Meng, Zewen Sun, Kangye Ji, Chen Tang, Jiajun Fan, Xinzhu Ma, Shu-Tao Xia, Zhi Wang, Wenwu Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add action-type aware model scheduling (ablation) | SP-VLA · LIBERO average | success rate (%): 71.52 -> 74.90 | Improved |
| add spatio-semantic token pruning (ablation) | SP-VLA · LIBERO average | speedup: 1.27 -> 1.35 | Improved |
| add Canny edge (contour) tokens to token pruning (ablation) | SP-VLA · LIBERO average | success rate (%): 23.93 -> 74.90 | Improved |
| preserve relative token positions and add edge tokens to baseline pruners (+R, +S) | FastVLM, VisionZip · LIBERO average | success rate (%): 74.60 -> 66.65 (FastVLM), 33.05 (VisionZip) | No clear effect |
| apply VLM token pruning (PruMerge + R) to OpenVLA | OpenVLA · LIBERO average | success rate (%): 74.60 -> 0.00 | Hurt |
| random token dropping | CogACT · SimplerEnv Visual Matching | success rate (%): 74.80 -> 20.90 | Hurt |
| apply SP-VLA to CogACT | CogACT · SimplerEnv Visual Aggregation | success rate (%): 61.30 -> 65.16 | Improved |
| apply SP-VLA to CogACT | CogACT · SimplerEnv WindowX | success rate (%): 51.30 -> 57.29 | Improved |
| apply SP-VLA to CogACT | CogACT · SimplerEnv Visual Matching | frequency (Hz): 3.77 -> 8.06 | Improved |
| apply SP-VLA to OpenVLA | OpenVLA · LIBERO | frequency (Hz): 3.18 -> 4.43 | Improved |
| increase action buffer size n from 6 to 8 (ablation) | SP-VLA · LIBERO-Object | success rate (%): 82.4 -> 68.4 | Hurt |
| lower tau from 0.5 to 3/8 (ablation) | SP-VLA · LIBERO-Object | success rate (%): 82.4 -> 64.4 | Hurt |
Later papers that compared with this method
1 later accepted paper reports + SP-VLA (128) as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA (NeurIPS 2026 · better on 4 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- SparseVLM — SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference (ICML 2025) · better on 2 of 2 shared results
- EfficientVLA — EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models (NeurIPS 2025) · better on 2 of 2 shared results
- VLA-Cache — VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching (NeurIPS 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.94)
- EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models (NeurIPS 2025 · similarity 0.93)
- VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching (NeurIPS 2025 · similarity 0.92)
- VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon (NeurIPS 2026 · similarity 0.91)
- ActionUNet: Improving Robustness of VLA Models with Efficient Multi-scale Fine-tuning (NeurIPS 2026 · similarity 0.91)
- VLA-REPLICA: A Low-Cost, Reproducible Benchmark for Real-World Evaluation of Vision-Language-Action Models (NeurIPS 2026 · similarity 0.91)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
