OmniDrive-R1 enhances autonomous driving VLMs with reinforcement-driven visual grounding

Researchers have introduced OmniDrive-R1, a novel framework for autonomous driving that integrates perception and reasoning using an interleaved Multi-modal Chain-of-Thought (iMCoT) mechanism. This approach addresses object hallucination issues common in Vision-Language Models by employing a reinforcement-driven visual grounding capability. The system utilizes a unique annotation-free training pipeline with the Clip-GRPO algorithm, which generates a grounding reward without requiring dense localization labels. Experiments show OmniDrive-R1 significantly boosts reasoning scores and accuracy compared to baseline models. AI

影响 Introduces a novel approach to improve VLM reliability in safety-critical autonomous driving applications.

排序理由 This is a research paper detailing a new model and methodology for autonomous driving.

在 arXiv cs.AI 阅读 →

• Clip-GRPO
• DriveLMM-o1
• OmniDrive-R1
• Qwen2.5VL-7B
• Vision-Language Models
• Zhenguo Zhang
• Chain-of-Thought

• 论文
• 模型发布

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