arXiv cs.AI· Xiyan Su, Finn Rasmus Sch\"afer, Hojun Lim, Frank Diermeyer, Johannes Betz, Markus Lienkamp·· 4 小时前AI 评分35
GuidedWAM：面向自动驾驶世界动作模型的无训练能量引导框架
GuidedWAM: Training-Free Energy Guidance for World Action Models in Autonomous Driving
AI 导读
GuidedWAM 是一个模型无关的框架，可在采样阶段将驾驶目标注入冻结的自动驾驶世界动作模型（WAM），无需重新训练。它把每个目标构建为可微能量，仅需前向访问冻结模型，并通过引导调度、逐能量归一化和速度有界修正保持稳定。在 NAVSIM v2 hard-split 上，三个公开冻结 WAM 均提升 1.8 到 3.6 EPDMS，每步采样时间增加不到 10%。
正文 · 原文
Abstract:World action models (WAMs) for autonomous driving couple the driving policy and the visual world model in a single generative network whose objectives are fixed post-training. Conditioning on a new driving objective requires retraining. We study energy guidance as a general mechanism and present GuidedWAM, a model-agnostic framework for injecting driving objectives into frozen WAMs at sampling time. Our framework constructs each objective as differentiable energies on the sampler's intermediate clean-action estimate, which needs only forward access to the frozen model, and is stabilized by sampling-time guidance scheduling, per-energy normalization, and velocity-bounded correction. On the NAVSIM v2 hard-split, guidance improves all three frozen, publicly released WAMs by 1.8 to 3.6 EPDMS, averaged over three sampler seeds, and adds less than 10% to the sampling time per step. These results position energy guidance as a practical post-training control channel for WAMs.
| Subjects: | Artificial Intelligence (cs.AI); Robotics (cs.RO) |
| Cite as: | arXiv:2611.07912 [cs.AI] |
| (or arXiv:2611.07912v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.07912 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xiyan Su [view email]
[v1]
Thu, 5 Nov 2026 12:27:38 UTC (963 KB)
来源：arXiv cs.AI · arxiv.org