NoRD：一种无需推理的数据高效视觉-语言-动作模型
NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning
对做自动驾驶VLA的研究者，这篇论文表明无需推理标注、仅用不到60%数据，通过Dr.GRPO强化学习微调SFT策略即可达到竞争性性能，且推理词元减少3倍，为数据高效、低延迟的端到端驾驶模型提供了可复用的训练范式。
原文依据 NoRD achieves competitive performance while being fine-tuned on atleast <60% of the data and no reasoning annotations, resulting in 3x fewer tokens.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉-语言-动作（VLA）模型通过用统一的端到端架构替代模块化流水线，正在推动自动驾驶的发展。当前的VLA面临两个挑战：（1）它们需要大量带有推理轨迹标注的数据集；（2）这些推理轨迹大幅增加了词元数量，抬高了训练和推理成本。我们提出NoRD（No Reasoning for Driving），一种数据高效且推理高效的VLA，同时解决这两个问题。与现有VLA相比，NoRD在仅使用不到60%的数据且无推理标注的情况下进行微调，仍达到有竞争力的性能，词元数量减少3倍。我们的方法应用强化学习（RL）对在一个小型、无推理数据集上训练的有监督微调（SFT）策略进行微调。然而，我们观察到标准RL算法——组相对策略优化（GRPO）——未能在此数据高效的SFT策略上带来显著提升。我们发现这一局限源于难度偏差，即GRPO中来自产生高方差轨迹的场景的奖励信号被不成比例地惩罚。NoRD通过引入Dr.GRPO克服了这一局限，该算法是近期为缓解大语言模型中的难度偏差而设计的。因此，NoRD在Waymo和NAVSIM上无需大规模数据集、推理或额外输入即可达到有竞争力的性能，实现了可扩展的数据高效训练和快速推理。
英文原文摘要 1410 字符
Vision-Language-Action (VLA) models are advancing autonomous driving by replacing modular pipelines with unified end-to-end architectures. Current VLAs face two challenges: (1) they require extensive datasets annotated with reasoning traces, and (2) these traces greatly increase token counts, inflating training and inference costs. We propose NoRD (No Reasoning for Driving), a data- and inference-efficient VLA that addresses both. Compared to existing VLAs, NoRD achieves competitive performance while being fine-tuned on atleast <60% of the data and no reasoning annotations, resulting in 3x fewer tokens. Our approach applies Reinforcement Learning (RL) to fine-tune a Supervised fine-tuning (SFT) policy trained on a small, reasoning-free dataset. However, we observe that the standard RL algorithm, Group Relative Policy Optimization (GRPO), fails to yield significant improvements over this data-efficient SFT policy. We find that this limitation stems from difficulty bias, which disproportionately penalizes reward signals from scenarios that produce high-variance rollouts within GRPO. NoRD overcomes this limitation by incorporating Dr.GRPO, a recent algorithm designed to mitigate difficulty bias in LLMs. As a result, NoRD achieves competitive performance on Waymo and NAVSIM without large datasets, reasoning or additional inputs, enabling scalable, data-efficient training, and fast inference.