GM trains scalable driving AI using simulation and VLA models for rare scenarios

General Motors is developing advanced AI systems to tackle the complex challenges of autonomous driving, particularly focusing on rare and unpredictable "long-tail" scenarios. They are employing a combination of large-scale simulation, reinforcement learning, and foundation models to train these systems. A key innovation is the use of Vision Language Action (VLA) models, which are fine-tuned for driving-specific tasks, and a "Dual Frequency VLA" approach to manage latency by separating high-level decision-making from immediate control actions. AI

排序理由 This is a sponsored article detailing research and development in AI for autonomous driving, rather than a product release or a new model.

在 IEEE Spectrum — AI 阅读 →

• General Motors
• IEEE Spectrum
• Vision Language Action

• 模型发布
• 产品
• 基础设施

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →