analiza aiqbench
LUNA-AD addresses critical bottlenecks in autonomous driving by combining a lightweight, uncertainty-aware architecture with lifelong learning to maintain strategic diversity without sacrificing inference speed. The system's tri-design effectively balances complex multimodal reasoning with efficient deployment, achieving state-of-the-art performance on nuPlan benchmarks while significantly reducing latency compared to traditional knowledge-driven frameworks. This approach offers a viable path forward for integrating large language models into safety-critical systems where computational overhead and static learning paradigms have previously been prohibitive.
LUNA-AD: Lightweight Uncertainty-Aware Language Model with Lifelong Learning for Autonomous Driving
arXiv:2606.08470v1 Announce Type: new Abstract: While large language models (LLMs) offer promising reasoning capabilities, their integration into safety-critical driving systems is hindered by limited reasoning diversity, high computational overhead, and static learning paradigms. To address these challenges, we propose LUNA-AD, a lightweight uncertainty-aware language model with lifelong learning for autonomous driving (AD). LUNA-AD features a tri-system architecture that reconciles complex multimodal behavioral reasoning, efficient deployment, and continual refinement. We design a multi-agent analytical system to generate uncertainty-aware decision-making demonstrations through diverse hypothesis exploration. A dual-head lightweight heuristic model is distilled to unify the inference of decision distributions and textual explanations while enabling efficient deployment. Furthermore, a reflection-driven lifelong learning mechanism operates on multimodal decision outputs and preserves strategic diversity, allowing for the refinement of candidate decisions and rationales via closed-loop feedback to enhance driving robustness. Extensive experiments on nuPlan benchmarks demonstrate that LUNA-AD achieves state-of-the-art success rates under both non-reactive and reactive modes, with drastically reduced inference latency compared to existing knowledge-driven AD frameworks.
Abstract:While large language models (LLMs) offer promising reasoning capabilities, their integration into safety-critical driving systems is hindered by limited reasoning diversity, high computational overhead, and static learning paradigms. To address these challenges, we propose LUNA-AD, a lightweight uncertainty-aware language model with lifelong learning for autonomous driving (AD). LUNA-AD features a tri-system architecture that reconciles complex multimodal behavioral reasoning, efficient deployment, and continual refinement. We design a multi-agent analytical system to generate uncertainty-aware decision-making demonstrations through diverse hypothesis exploration. A dual-head lightweight heuristic model is distilled to unify the inference of decision distributions and textual explanations while enabling efficient deployment. Furthermore, a reflection-driven lifelong learning mechanism operates on multimodal decision outputs and preserves strategic diversity, allowing for the refinement of candidate decisions and rationales via closed-loop feedback to enhance driving robustness. Extensive experiments on nuPlan benchmarks demonstrate that LUNA-AD achieves state-of-the-art success rates under both non-reactive and reactive modes, with drastically reduced inference latency compared to existing knowledge-driven AD frameworks.
Robotics (cs.RO)
arXiv:2606.08470 [cs.RO]
(or arXiv:2606.08470v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.08470 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 252 cuvinte
Subiecte conexe
[AINews] OpenAI DevDay 2026: Dots, 6.1 Sol, Ultrafast, Decisions API, Agents API, Spaces, Marketplace, and 1.2 Billion ChatGPT WAU
Latent Space · 30 sept. 2026
Dynamic Distribution-Aware Uncertainty Tracking in Vision-Language Representation Learning
arXiv · 20 aug. 2026
ChainUQ: Reasoning Consistency-Aware Uncertainty Quantification for Large Language Models
arXiv · 24 sept. 2026
Towards Human-like Physical Intelligence: Lifelong Vision-Language-Action Learning for Robotic Manipulation
arXiv · 22 iul. 2026