mlsanityData to 14 Oct 2026

Hidden Biases of End-to-End Driving Models

arXiv 2306.07957 ↗First posted 13 Jun 2023cs.CV cs.AI cs.LGcited by 96

On the map: End-to-end autonomous driving → End-to-end system architecture

Abstract (arXiv)

End-to-end driving systems have recently made rapid progress, in particular on CARLA. Independent of their major contribution, they introduce changes to minor system components. Consequently, the source of improvements is unclear. We identify two biases that recur in nearly all state-of-the-art methods and are critical for the observed progress on CARLA: (1) lateral recovery via a strong inductive bias towards target point following, and (2) longitudinal averaging of multimodal waypoint predictions for slowing down. We investigate the drawbacks of these biases and identify principled alternatives. By incorporating our insights, we develop TF++, a simple end-to-end method that ranks first on the Longest6 and LAV benchmarks, gaining 11 driving score over the best prior work on Longest6.

Key facts. Cited by 96 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 31 of its 44 references that are in the map. First posted on arXiv 13 Jun 2023.

Builds on

31 of 44 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023Think Twice before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving10 May
  2. 2022Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline16 Jun
  3. 2022TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving31 May
  4. 2021Multi-Modal Fusion Transformer for End-to-End Autonomous Driving19 Apr

Other references · 27

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016Layer Normalizationcited by 5,669
  3. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
  4. 2016Gaussian Error Linear Units (GELUs)cited by 3,502
  5. 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
  6. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  7. 2016End to End Learning for Self-Driving Carscited by 1,009
  8. 2020Designing Network Design Spacescited by 960
  9. 2019On the Convergence of Adam and Beyondcited by 841
  10. 2017End-to-end Driving via Conditional Imitation Learningcited by 485
  11. 2018ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worstcited by 397
  12. 2019Learning by Cheatingcited by 383
15 more
  1. 2019Exploring the Limitations of Behavior Cloning for Autonomous Drivingcited by 346
  2. 2021MP3: A Unified Model to Map, Perceive, Predict and Plancited by 212
  3. 2022Safety-Enhanced Autonomous Driving Using Interpretable Sensor Fusion Transformercited by 212
  4. 2021End-to-End Urban Driving by Imitating a Reinforcement Learning Coachcited by 194
  5. 2021NEAT: Neural Attention Fields for End-to-End Autonomous Drivingcited by 174
  6. 2022Model-Based Imitation Learning for Urban Drivingcited by 150
  7. 2019End-to-End Model-Free Reinforcement Learning for Urban Driving using Implicit Affordancescited by 149
  8. 2022Learning from All Vehiclescited by 135
  9. 2021Learning to drive from a world on railscited by 114
  10. 2022PlanT: Explainable Planning Transformers via Object-Level Representationscited by 105
  11. 2023ReasonNet: End-to-End Driving with Temporal and Global Reasoningcited by 93
  12. 2021VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehiclescited by 81
  13. 2020Fighting Copycat Agents in Behavioral Cloning from Observation Historiescited by 71
  14. 2018Conditional Affordance Learning for Driving in Urban Environmentscited by 70
  15. 2020Label Efficient Visual Abstractions for Autonomous Drivingcited by 33

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 5; 2024: 24; 2025: 35; 2026: 32.

96 papers.

Sort:

By area:

  1. 2023DriveLM: Driving with Graph Visual Question Answeringcited by 536
  2. 2023End-to-end Autonomous Driving: Challenges and Frontierscited by 490
  3. 2024NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarkingcited by 354
  4. 2023LMDrive: Closed-Loop End-to-End Driving with Large Language Modelscited by 305
  5. 2025AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuningcited by 260
  6. 2024Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Drivingcited by 252
  7. 2025ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generationcited by 178
  8. 2024Enhancing End-to-End Autonomous Driving with Latent World Modelcited by 166
  9. 2025SimLingo: Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignmentcited by 150
  10. 2025Pseudo-Simulation for Autonomous Drivingcited by 150
  11. 2023DriveAdapter: Breaking the Coupling Barrier of Perception and Planning in End-to-End Autonomous Drivingcited by 144
  12. 2025DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Drivingcited by 126
38 more
  1. 2023Recent Advancements in End-to-End Autonomous Driving using Deep Learning: A Surveycited by 103
  2. 2024DrivingGPT: Unifying Driving World Modeling and Planning with Multi-modal Autoregressive Transformerscited by 96
  3. 2025DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planningcited by 94
  4. 2024RAG-Driver: Generalisable Driving Explanations with Retrieval-Augmented In-Context Learning in Multi-Modal Large Language Modelcited by 92
  5. 2024VLM-AD: End-to-End Autonomous Driving through Vision-Language Model Supervisioncited by 86
  6. 2025DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Drivingcited by 86
  7. 2025A Survey on Vision-Language-Action Models for Autonomous Drivingcited by 72
  8. 2024DriveCoT: Integrating Chain-of-Thought Reasoning with End-to-End Drivingcited by 67
  9. 2025Robust Autonomy Emerges from Self-Playcited by 67
  10. 2024CarLLaVA: Vision language models for camera-only closed-loop drivingcited by 57
  11. 2024Think2Drive: Efficient Reinforcement Learning by Thinking in Latent World Model for Quasi-Realistic Autonomous Driving (in CARLA-v2)cited by 52
  12. 2025Centaur: Robust End-to-End Autonomous Driving with Test-Time Trainingcited by 42
  13. 2025Hydra-NeXt: Robust Closed-Loop Driving with Open-Loop Trainingcited by 39
  14. 2024Can Vehicle Motion Planning Generalize to Realistic Long-tail Scenarios?cited by 38
  15. 2026SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Drivingcited by 38
  16. 2025HiP-AD: Hierarchical and Multi-Granularity Planning with Deformable Attention for Autonomous Driving in a Single Decodercited by 35
  17. 2025CaRL: Learning Scalable Planning Policies with Simple Rewardscited by 35
  18. 2024Hidden Biases of End-to-End Driving Datasetscited by 30
  19. 2024Delving into Multi-modal Multi-task Foundation Models for Road Scene Understanding: From Learning Paradigm Perspectivescited by 23
  20. 2025LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Drivingcited by 23
  21. 2024Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Modelscited by 22
  22. 2024LangProp: A code optimization framework using Large Language Models applied to drivingcited by 18
  23. 2025BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Drivingcited by 18
  24. 2026Fail2Drive: Benchmarking Closed-Loop Driving Generalizationcited by 17
  25. 2024ActiveAD: Planning-Oriented Active Learning for End-to-End Autonomous Drivingcited by 16
  26. 2025GaussianFusion: Gaussian-Based Multi-Sensor Fusion for End-to-End Autonomous Drivingcited by 16
  27. 2025ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Drivingcited by 14
  28. 2024Autonomous Driving with Spiking Neural Networkscited by 13
  29. 2026DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Drivingcited by 11
  30. 2024Efficient Domain Augmentation for Autonomous Driving Testing Using Diffusion Modelscited by 10
  31. 2025DriveQA: Passing the Driving Knowledge Testcited by 10
  32. 2025DiffRefiner: Coarse to Fine Trajectory Planning via Diffusion Refinement with Semantic Interaction for End to End Autonomous Drivingcited by 10
  33. 2026MindDriver: Introducing Progressive Multimodal Reasoning for Autonomous Drivingcited by 10
  34. 2025DiffE2E: Rethinking End-to-End Driving with a Hybrid Action Diffusion and Supervised Policycited by 9
  35. 2025ETA: Efficiency through Thinking Ahead, A Dual Approach to Self-Driving with Large Modelscited by 9
  36. 2026AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture-of-Transformers for End-to-End Autonomous Drivingcited by 9
  37. 2024Analysis of a Modular Autonomous Driving Architecture: The Top Submission to CARLA Leaderboard 2.0 Challengecited by 8
  38. 2025PlanT 2.0: Exposing Biases and Structural Flaws in Closed-Loop Drivingcited by 8

50 most cited of 96 All 96 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Addressing the Waypoint-Action Gap in End-to-End Autonomous Driving via Vehicle Motion Models
  2. 2024Data Scaling Laws for Imitation Learning-Based End-to-End Autonomous Driving
  3. 2025WOD-E2E: Waymo Open Dataset for End-to-End Driving in Challenging Long-tail Scenarios
  4. 2026Multimodal Action Diffusion for Robust End-to-End Autonomous Driving
  5. 2025AnchDrive: Bootstrapping Diffusion Policies with Hybrid Trajectory Anchors for End-to-End Driving
  6. 2026Driving on Memory
  7. 2025Closing the Loop: Motion Prediction Models beyond Open-Loop Benchmarks
  8. 2026Odyssey: A Closed-Loop Benchmark for Long-Horizon Real-World Driving with Explicit Navigation Routes