English(EN)Understanding Autonomous Driving Datasets by Describing Differences between Image Subsets in Natural Language
Qwen-Drive-1.0 集成了 3D 感知、VQA 和运动规划，用于自动驾驶
作者PulseAugur 编辑部·[7 个来源]·
Qwen 推出了 Qwen-Drive-1.0，一个专为自动驾驶设计的视觉语言基础模型。该模型通过利用预训练的 VLM 架构和外部模块，统一了 3D 感知、视觉问答和运动规划。它包含一个鸟瞰视角感知头，用于 3D 对象检测和语义占用预测等任务，以及一个用于生成未来轨迹的规划专家。该模型采用分阶段的配方进行训练，结合了驾驶监督和通用视觉语言数据，以保持广泛的能力。
AI
We introduce Qwen-Drive-1.0, the first vision-language foundation model for autonomous driving that unifies 3D perception and visual question answering at the pretraining stage and further extends to motion planning, while keeping the pretrained VLM architecture entirely untouche…
arXiv:2505.00284v3 Announce Type: replace-cross Abstract: Rapid advances in vision-language models (VLMs) have generated growing interest in their application to autonomous driving. A prevailing assumption is that successive VLM generations will continually improve driving perfor…
arXiv cs.CL
TIER_1English(EN)·Julian Truetsch, Felix Hauser, Christoph Stiller, Frank Bieder·
arXiv:2609.03677v1 Announce Type: cross Abstract: Understanding the composition of large-scale autonomous driving datasets is essential for safety, robustness, and reliable operation across domains. For example, domain shift between locations could lead to the operating environme…
Understanding the composition of large-scale autonomous driving datasets is essential for safety, robustness, and reliable operation across domains. For example, domain shift between locations could lead to the operating environment being misaligned with the training data, result…
Understanding the composition of large-scale autonomous driving datasets is essential for safety, robustness, and reliable operation across domains. For example, domain shift between locations could lead to the operating environment being misaligned with the training data, result…
Qwen-Drive-1.0 is a vision-language foundation model for autonomous driving that unifies 3D perception, visual question answering, and motion planning via shared representations and staged training.
arXiv:2609.00111v1 Announce Type: new Abstract: We present Qwen-Drive-1.0, an initial step towards a vision-language foundation model for autonomous driving. Qwen-Drive-1.0 retains the architecture of the pretrained vision-language model (VLM) and integrates 3D perception, visual…
