Hint-AD: Holistically Aligned Interpretability in End-to-End Autonomous Driving
This work addresses the problem of human-AI trust in autonomous driving by providing more grounded and interpretable language outputs, though it appears incremental as it builds on prior interpretability methods.
The authors tackled the interpretability challenge in end-to-end autonomous driving by introducing Hint-AD, a system that generates language aligned with intermediate perception-prediction-planning outputs, achieving state-of-the-art results in driving explanation, 3D dense captioning, and command prediction tasks.
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.