Research map / Vision-language and multimodal / Robot action and navigation
SpikeVLA: Vision-Language-Action Models with Spiking Neural Networks
Ruiqi Song, Dujun Nie, Siyu Teng, Baiyong Ding, Xiaotong Zhang, Dong Li, Chenming Zhang, Yuchen Li, Hangbin Wu, Long Chen
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 |
|---|---|---|---|
| replace dense ANN VLA with spiking SpikeVLA | NaVILA · R2R-CE Val-Unseen | energy (J): 141.25 -> 49.09 | Improved |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA · R2R-CE Val-Unseen | SR: 53.9 -> 53.3 | No clear effect |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA · R2R-CE Val-Unseen | SPL: 49.3 -> 47.9 | Hurt |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA · R2R-CE Val-Unseen | OS: 61.5 -> 63.4 | Improved |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA · RxR-CE Val-Unseen | SR: 52.3 -> 51.9 | No clear effect |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA-R · VLN-CE-Isaac | SR: 36.5 -> 32.7 | Hurt |
| replace dense ANN VLA with spiking SpikeVLA | NaVILA-R · VLN-CE-Isaac | energy (J): 141.25 -> 44.23 | Improved |
| use spiking action policy instead of ANN policy | NaVILA low-level policy · closed-loop locomotion | linear velocity error: 0.23 -> 0.42 | Hurt |
| use spiking action policy instead of ANN policy | NaVILA low-level policy · closed-loop locomotion | angular velocity error: 0.38 -> 0.29 | Improved |
| use spiking action policy instead of ANN policy | NaVILA low-level policy · closed-loop locomotion | energy (uJ): 5.80 -> 0.31 | Improved |
| INT4 quantization of ANN VLA (ablation) | NaVILA · R2R Val-Unseen | SR: 53.9 -> 48.2 | Hurt |
| INT4 quantization of ANN VLA (ablation) | NaVILA · R2R Val-Unseen | energy (J): 141.25 -> 72.49 | Improved |
Closest papers by meaning
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.90)
- EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models (NeurIPS 2025 · similarity 0.89)
- NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models (NeurIPS 2026 · similarity 0.88)
- CogVLA: Cognition-Aligned Vision-Language-Action Models via Instruction-Driven Routing & Sparsification (NeurIPS 2025 · similarity 0.88)
- D-VLA: A High-Concurrency Distributed Asynchronous RL Framework for Vision-Language-Action Models (NeurIPS 2026 · similarity 0.88)
- VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching (NeurIPS 2025 · similarity 0.88)
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.
