Hint-AD：端到端自动驾驶中的整体对齐可解释性
Hint-AD: Holistically Aligned Interpretability in End-to-End Autonomous Driving
AI总结：
针对端到端自动驾驶自然语言解释未锚定中间输出的问题，提出Hint-AD集成系统，通过整体token mixer对齐感知-预测-规划结果，并发布Nu-X数据集。
摘要：
自动驾驶（AD）中的端到端架构在可解释性方面面临重大挑战，这阻碍了人类与AI之间的信任。人类友好的自然语言已被用于驾驶解释和3D描述等任务。然而，以往工作主要聚焦于陈述式可解释性范式，其中自然语言解释并未锚定到AD系统的中间输出，导致解释仅停留在陈述层面。相比之下，对齐式可解释性在语言与AD系统的中间输出之间建立联系。本文提出Hint-AD，这是一个集成的AD-语言系统，可生成与AD模型的整体感知-预测-规划输出相对齐的语言。通过融合中间输出以及用于有效特征适配的整体token mixer子网络，Hint-AD取得了理想的准确率，并在驾驶解释、3D密集描述和指令预测等驾驶语言任务上达到最先进结果。为促进在nuScenes上进一步研究驾驶解释任务，我们还引入了一个人工标注数据集Nu-X。代码、数据集和模型将公开发布。
英文摘要：
End-to-end architectures in autonomous driving (AD) face a significant challenge in interpretability, impeding human-AI trust. Human-friendly natural language has been explored for tasks such as driving explanation and 3D captioning. However, previous works primarily focused on the paradigm of declarative interpretability, where the natural language interpretations are not grounded in the intermediate outputs of AD systems, making the interpretations only declarative. In contrast, aligned interpretability establishes a connection between language and the intermediate outputs of AD systems. Here we introduce Hint-AD, an integrated AD-language system that generates language aligned with the holistic perception-prediction-planning outputs of the AD model. By incorporating the intermediate outputs and a holistic token mixer sub-network for effective feature adaptation, Hint-AD achieves desirable accuracy, achieving state-of-the-art results in driving language tasks including driving explanation, 3D dense captioning, and command prediction. To facilitate further study on driving explanation task on nuScenes, we also introduce a human-labeled dataset, Nu-X. Codes, dataset, and models will be publicly available.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
