A Super-Aligned Driving Generalist Is Your Cockpit

Hao LU, Jiaqi Tang, Jiyao Wang, Yunfan LU, Qingyong Hu, Yin Wang, Tianxin Xie, Yuting Zhang, Yunpeng Zhang, Bin Huang, Dengbo He, Shuiguang Deng, Hao Chen, Ying-Cong Chen

Keywords: Driving Cockpit; Super alined; Driving Generalist

Abstract: The intelligent driving cockpit, an important part of intelligent driving, needs to match different users' comfort, interaction, and safety needs. This paper aims to build a \textbf{s}uper-\textbf{a}ligned and \textbf{ge}neralist \textbf{dr}iving agent, \textbf{sage deer}. Sage Deer achieves two highlights: (1) Super alignment: It achieves different reactions according to different people's preferences and biases. (2) Generalist: It can understand the user's physiological indicators, facial emotions, hand movements, body movements, driving scenarios, and behavioral decisions. (3) Multimodal: He can understand RGB, NIR, and depth video to build more robust perception, understanding, and reasoning. To achieve the above requirements, we design retrieval-enhanced multimodal frameworks. We collected multiple data sets and built a large-scale benchmark. This benchmark measures the sage deer's perceptual decision-making ability and the super alignment's accuracy.

Primary Area: applications to robotics, autonomy, planning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 174