Less is More: Lean yet Powerful Vision-Language Model for Autonomous Driving
In this work, we reconceptualize autonomous driving as a generalized language problem and formulate the trajectory planning task as next waypoint prediction. We introduce Max-V1, a novel framework for one-stage end-to-end autonomous driving, named in tribute to the renowned Dutch racing driver Max Verstappen. Our framework presents a single-pass generation paradigm that aligns with the inherent sequentiality of driving. This approach leverages the generative capacity of the Vision-Language Model (VLM) to enable end-to-end trajectory prediction directly from front-view camera input. The efficacy of this method is underpinned by a principled supervision strategy derived from statistical modeling. This provides a well-defined learning objective, which makes the framework highly amenable to mastering complex driving policies through imitation learning from large-scale expert demonstrations. Empirically, our method achieves state-of-the-art performance on the nuScenes dataset, delivering an overall improvement of over 30% compared to prior baselines. Furthermore, it exhibits superior generalization performance on cross-domain datasets acquired from diverse vehicles, demonstrating notable potential for cross-vehicle robustness and adaptability. With these empirical strengths, this work introduces a model that enables fundamental driving behaviors, laying the foundation for the development of more capable self-driving agents. Code will be available upon publication.
Code (0)
등록된 구현이 없습니다.
Tasks
Trajectory PredictionTrajectory PlanningAutonomous DrivingSimilar Papers 제목 키워드 기반
PIP: Detecting Adversarial Examples in Large Vision-Language Models via Attention Patterns of Irrelevant Probe Questions
Large Vision-Language Models (LVLMs) have demonstrated their powerful multimodal capabilities. However, they also face serious safety problems, as adversaries can induce robustness issues in LVLMs through the use of well…
INN: A Method Identifying Clean-annotated Samples via Consistency Effect in Deep Neural Networks
In many classification problems, collecting massive clean-annotated data is not easy, and thus a lot of researches have been done to handle data with noisy labels. Most recent state-of-art solutions for noisy label probl…
MemorizationVALD: Multi-Stage Vision Attack Detection for Efficient LVLM Defense
Large Vision-Language Models (LVLMs) can be vulnerable to adversarial images that subtly bias their outputs toward plausible yet incorrect responses. We introduce a general, efficient, and training-free defense that comb…
CLIPCleaner: Cleaning Noisy Labels with CLIP
Learning with Noisy labels (LNL) poses a significant challenge for the Machine Learning community. Some of the most widely used approaches that select as clean samples for which the model itself (the in-training model) h…
Learning with noisy labelsImproving Chemical Understanding of LLMs via SMILES Parsing
Large language models (LLMs) are increasingly recognized as powerful tools for scientific discovery, particularly in molecular science. A fundamental requirement for these models is the ability to accurately understand m…
Graph Matchingscientific discovery