无需微调，让预训练视觉语言动作模型在真实机器人上直接可用
Towards Deploying VLA without Fine-Tuning: Plug-and-Play Inference-Time VLA Policy Steering via Embodied Evolutionary Diffusion
- 推理时通过进化扩散动态调整策略，实现即插即用
- 六项真实任务成功率显著提升，跨机器人形态零样本泛化
- 适合希望快速部署现有模型的工业界研究者
视觉语言动作（VLA）模型在现实世界机器人操作中展现出巨大潜力。然而，预训练的VLA策略在下游部署时仍面临性能大幅下降的问题。尽管微调可缓解此问题，但其依赖昂贵的示范数据收集和高计算成本，在实际场景中难以应用。本文提出VLA-Pilot，一种无需额外微调或数据收集的即插即用推理时策略调控方法，支持预训练VLA的零样本部署。我们在两个不同机器人本体上评估了六项真实世界的下游操作任务，涵盖分布内与分布外情形。实验结果表明，VLA-Pilot显著提升了现成预训练VLA策略的成功率，实现了对多样化任务和本体的鲁棒零样本泛化。实验视频与代码见：https://rip4kobe.github.io/vla-pilot/
原文摘要 · Abstract (English)
Vision-Language-Action (VLA) models have demonstrated significant potential in real-world robotic manipulation. However, pre-trained VLA policies still suffer from substantial performance degradation during downstream deployment. Although fine-tuning can mitigate this issue, its reliance on costly demonstration collection and intensive computation makes it impractical in real-world settings. In this work, we introduce VLA-Pilot, a plug-and-play inference-time policy steering method for zero-shot deployment of pre-trained VLA without any additional fine-tuning or data collection. We evaluate VLA-Pilot on six real-world downstream manipulation tasks across two distinct robotic embodiments, encompassing both in-distribution and out-of-distribution scenarios. Experimental results demonstrate that VLA-Pilot substantially boosts the success rates of off-the-shelf pre-trained VLA policies, enabling robust zero-shot generalization to diverse tasks and embodiments. Experimental videos and code are available at: https://rip4kobe.github.io/vla-pilot/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。