https://auto-driving-blog.vercel.app/tags/diffusiondrivev2/ Recent content in DiffusionDriveV2 on Elon's AD Insight https://auto-driving-blog.vercel.app/images/share.png https://auto-driving-blog.vercel.app/images/share.png Hugo zh-cn Mon, 20 Jul 2026 00:00:00 +0000 https://auto-driving-blog.vercel.app/posts/code/diffusiondrivev2%E4%BB%A3%E7%A0%81%E8%AE%B2%E8%A7%A3/ Mon, 20 Jul 2026 00:00:00 +0000 https://auto-driving-blog.vercel.app/posts/code/diffusiondrivev2%E4%BB%A3%E7%A0%81%E8%AE%B2%E8%A7%A3/ 「DiffusionDriveV2 在 DiffusionDrive 的 anchor 截断扩散之上，补了一套 GRPO 强化学习微调：用 scale-adaptive 乘性噪声做探索、Intra-Anchor GRPO 保住多模态不坍缩、Inter-Anchor Truncated GRPO 用碰撞惩罚把低质量轨迹压下去，最后加两级 Mode Selector 精排，在 NAVSIM v1 上冲到 91.2 PDMS。本文基于论文 Algorithm 还原成逐文件逐函数伪代码，从 cold start 权重加载写到 rollout → advantage → loss 的完整 RL 训练循环。」