# CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention arXiv: https://arxiv.org/abs/2603.18561 ยท HTML: https://mlsanity.com/paper/2603.18561/ ## Key facts - Cited by 7 papers in the map (data through 14 Oct 2026). - Placed in End-to-end system architecture, one of 9 directions in End-to-end autonomous driving. - Builds on 30 of its 36 references that are in the map. - First posted on arXiv 19 Mar 2026. ## Placed in - End-to-end autonomous driving / [End-to-end system architecture](https://mlsanity.com/driving/end-to-end-system-architecture/) ## Builds on (most cited first, 30 of 30) - [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019) - [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020) - [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015) - [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022) - [VAD: Vectorized Scene Representation for Efficient Autonomous Driving](https://mlsanity.com/paper/2303.12077/) (2023) - [DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models](https://mlsanity.com/paper/2402.12289/) (2024) - [ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst](https://mlsanity.com/paper/1812.03079/) (2018) - [Learning by Cheating](https://mlsanity.com/paper/1912.12294/) (2019) - [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022) - [A Survey on Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2402.00253/) (2024) - [NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking](https://mlsanity.com/paper/2406.15349/) (2024) - [DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving](https://mlsanity.com/paper/2411.15139/) (2024) - [Causal Confusion in Imitation Learning](https://mlsanity.com/paper/1905.11979/) (2019) - [Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving](https://mlsanity.com/paper/2406.03877/) (2024) - [Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving?](https://mlsanity.com/paper/2312.03031/) (2023) - [SparseDrive: End-to-End Autonomous Driving via Sparse Scene Representation](https://mlsanity.com/paper/2405.19620/) (2024) - [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024) - [ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation](https://mlsanity.com/paper/2503.19755/) (2025) - [Senna: Bridging Large Vision-Language Models and End-to-End Autonomous Driving](https://mlsanity.com/paper/2410.22313/) (2024) - [OmniDrive: A Holistic Vision-Language Dataset for Autonomous Driving with Counterfactual Reasoning](https://mlsanity.com/paper/2504.04348/) (2025) - [Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives](https://mlsanity.com/paper/2501.04003/) (2025) - [Visual Commonsense R-CNN](https://mlsanity.com/paper/2002.12204/) (2020) - [A Survey on Vision-Language-Action Models for Autonomous Driving](https://mlsanity.com/paper/2506.24044/) (2025) - [PPAD: Iterative Interactions of Prediction and Planning for End-to-end Autonomous Driving](https://mlsanity.com/paper/2311.08100/) (2023) - [PointSSC: A Cooperative Vehicle-Infrastructure Point Cloud Benchmark for Semantic Scene Completion](https://mlsanity.com/paper/2309.12708/) (2023) - [Towards Context-Aware Emotion Recognition Debiasing from a Causal Demystification Perspective via De-confounded Training](https://mlsanity.com/paper/2407.04963/) (2024) - [Safety Implications of Explainable Artificial Intelligence in End-to-End Autonomous Driving](https://mlsanity.com/paper/2403.12176/) (2024) - [PAPL-SLAM: Principal Axis-Anchored Monocular Point-Line SLAM](https://mlsanity.com/paper/2410.12324/) (2024) - [Decoupling Scene Perception and Ego Status: A Multi-Context Fusion Approach for Enhanced Generalization in End-to-End Autonomous Driving](https://mlsanity.com/paper/2511.13079/) (2025) - [Learning Spatial-Aware Manipulation Ordering](https://mlsanity.com/paper/2510.25138/) (2025) ## Built on by (most cited first, 7 of 7) - [EponaV2: Driving World Model with Comprehensive Future Reasoning](https://mlsanity.com/paper/2605.14696/) (2026) - [The DAWN of World-Action Interactive Models](https://mlsanity.com/paper/2605.11550/) (2026) - [IDOL: Inverse-Dynamics-Guided Future Prediction for End-to-End Autonomous Driving](https://mlsanity.com/paper/2605.31476/) (2026) - [Neuro-Symbolic Drive: Rule-Grounded Faithful Reasoning for Driving VLAs](https://mlsanity.com/paper/2606.23938/) (2026) - [DINO-VO: Learning Where to Focus for Enhanced State Estimation](https://mlsanity.com/paper/2604.04055/) (2026) - [Physics-Grounded Causal Auditing of End-to-End Driving Planners](https://mlsanity.com/paper/2606.14438/) (2026) - [PrismAD: Decoupled Planning via Semantic Mixture-of-Planners for End-to-End Autonomous Driving](https://mlsanity.com/paper/2607.10336/) (2026) ## Abstract (arXiv) Planning-oriented end-to-end driving models show great promise, yet they fundamentally learn statistical correlations instead of true causal relationships. This vulnerability leads to causal confusion, where models exploit dataset biases as shortcuts, critically harming their reliability and safety in complex scenarios. To address this, we introduce CausalVAD, a de-confounding training framework that leverages causal intervention. At its core, we design the sparse causal intervention scheme (SCIS), a lightweight, plug-and-play module to instantiate the backdoor adjustment theory in neural networks. SCIS constructs a dictionary of prototypes representing latent driving contexts. It then uses this dictionary to intervene on the model's sparse vectorized queries. This step actively eliminates spurious associations induced by confounders, thereby eliminating spurious factors from the representations for downstream tasks. Extensive experiments on benchmarks like nuScenes show CausalVAD achieves state-of-the-art planning accuracy and safety. Furthermore, our method demonstrates superior robustness against both data bias and noisy scenarios configured to induce causal confusion.