We present RISE (Roadside Infrastructure Sequence Understanding and Evaluation), a framework spanning metric 3D tracking and structured vision-language reasoning in roadside sequences. For metric tracking, our image-only method combines SAM3 video identities with calibration-guided mask agreement for multi-view identit...
Yan-Bo Jiang, Haotian Zheng, Jia-Hao Wang et al.· 0 citations
Driving decisions depend on both where to focus and how to act on what is seen. Effective driving reasoning must establish which objects matter, where they are, and how they inform the intended action. Text-based rationales can describe a driving response while leaving its correspondence to specific visual evidence imp...
Zhi-Yuan Liu, Ze-Hong Ke, Yuan-Xin Tian et al.· 0 citations
Safety is a fundamental requirement for autonomous driving, yet existing end-to-end driving models still lack explicit risk-aware learning capacities. Existing rule-based risk models provide interpretable safety priors, yet their absolute risk scores depend on handcrafted functions, coefficients, and thresholds. Learni...
Yuan-Xin Tian, Zhi-Yuan Liu, Jin-Hao Li et al.· 0 citations
Vault is a framework that couples one-step latent generation with sample-based reinforcement learning guidance, and achieves state-of-the-art performance at real-time inference cost, reaching state-of-the-art performance at real-time inference cost.
Yi-Ning Xing, Ze-Hong Ke, Zhi-Yuan Liu et al.· 2 citations
Autonomous driving requires more than recognizing what is present in a scene: a planner must determine how road structure, surrounding agents, and their motion states should influence a future maneuver. Existing learning-based planners can capture these influences through latent scene features and trajectory decoders,...
Zhi-Yuan Liu, Yuan-Xin Tian, Ze-Hong Ke et al.· 0 citations
Vehicle-to-Everything (V2X) cooperation enables beyond-line-of-sight perception, mitigating occlusions in single-vehicle sensing. However, existing V2X benchmarks provide limited support for closed-loop evaluation and language-grounded supervision, hindering the development of vision-language models (VLMs) for end-to-e...
DRIFT, a fixed-depth planner that combines one-step drifting in a compact trajectory latent space with scene-aware proposal aggregation, is presented, showing that one-step latent proposal generation and direct aggregation provide an efficient design for multi-hypothesis motion planning.
Yi-Ning Xing, Zhi-Yuan Liu, Ze-Hong Ke et al.· arXiv.org· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.