端到端驾驶论文趋势 2023–2026

主要结论

• 端到端驾驶论文从 2023 年的 33 篇增长到 2026 年的 58 篇，三年间扩大到约 1.8 倍。
• CVPR 是发端到端驾驶论文最多的会议，2023–2026 年共 71 篇，占四个会议总量的 44%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 11 篇。
• 在「自动驾驶」的 5 个主题里，它的 2026 年论文数排第 5；同组增长最快的运动规划是 2.4 倍，端到端驾驶是 1.8 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026AlignDrive: Aligned Lateral-Longitudinal Planning for End-to-End Autonomous Driving
• ICML 2026Hide&Seek: Learning to Explain in an End-to-End Differentiable Network
• ICLR 2026RAP: 3D Rasterization Augmented End-to-End Planning
• CVPR 2026Bias at the End of the Score
• NeurIPS 2025Embodied Cognition Augmented End2End Autonomous Driving
• ICML 2025GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance
• ICLR 2025MGMapNet: Multi-Granularity Representation Learning for End-to-End Vectorized HD Map Construction
• CVPR 2025Bridging Past and Future: End-to-End Autonomous Driving with Historical Prediction and Planning
• NeurIPS 2024Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving
• ICML 2024Learning Exceptional Subgroups by End-to-End Maximizing KL-Divergence
• ICLR 2024DV-3DLane: End-to-end Multi-modal 3D Lane Detection with Dual-view Representation
• CVPR 2024HIMap: HybrId Representation Learning for End-to-end Vectorized HD Map Construction
• NeurIPS 2023[Re] End-to-end Algorithm Synthesis with Recurrent Networks: Logical Extrapolation Without Overthinking
• ICML 2023End-to-end Differentiable Clustering with Associative Memories
• ICLR 2023E3Bind: An End-to-End Equivariant Network for Protein-Ligand Docking
• CVPR 2023End-to-End Vectorized HD-Map Construction With Piecewise Bezier Curve
• NeurIPS 2026On the Existence of Uniformly Optimal Policies in MDPs under Epistemic Uncertainty
• ICML 2026RoCA: Robust Cross-Domain End-to-End Autonomous Driving
• ICLR 2026THE END OF MANUAL DECODING: TOWARDS TRULY END-TO-END LANGUAGE MODELS
• CVPR 2026LEAD Minimizing Learner-Expert Asymmetry in End-to-End Driving

全部端到端驾驶论文

共 161 篇，第 2 / 4 页

51. NeurIPS 2026Driving on Memory
52. NeurIPS 2026HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models
53. NeurIPS 2026LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
54. NeurIPS 2026On the Existence of Uniformly Optimal Policies in MDPs under Epistemic Uncertainty
55. NeurIPS 2026TopoCurve: Geometry-Aware Topology Reasoning via Bézier Curves in Autonomous Driving
56. NeurIPS 2026Towards Generalizable Data Valuation: Learning an End-to-end Deep Model to Compute Shapley Values
57. NeurIPS 2026Trajectory Evaluation via Rollout-Free World Model for End-to-end Autonomous Driving
58. NeurIPS 2026TwinFlux: One-Step Discrete-Continuous Flow for End-to-End Autonomous Driving
59. CVPR 2025Bridging Past and Future: End-to-End Autonomous Driving with Historical Prediction and Planning
60. CVPR 2025CryptoFace: End-to-End Encrypted Face Recognition
61. CVPR 2025DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving
62. CVPR 2025Distilling Multi-modal Large Language Models for Autonomous Driving
63. CVPR 2025Don't Shake the Wheel: Momentum-Aware Planning in End-to-End Autonomous Driving
64. CVPR 2025DriveGPT4-V2: Harnessing Large Language Model Capabilities for Enhanced Closed-Loop Autonomous Driving
65. CVPR 2025Driving by the Rules: A Benchmark for Integrating Traffic Sign Regulations into Vectorized HD Map
66. CVPR 2025End-to-End HOI Reconstruction Transformer with Graph-based Encoding
67. CVPR 2025End-to-End Implicit Neural Representations for Classification
68. CVPR 2025GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving
69. CVPR 2025RC-AutoCalib: An End-to-End Radar-Camera Automatic Calibration Network
70. CVPR 2025S4-Driver: Scalable Self-Supervised Driving Multimodal Large Language Model with Spatio-Temporal Visual Representation
71. ICLR 2025DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving
72. ICLR 2025Enhancing End-to-End Autonomous Driving with Latent World Model
73. ICLR 2025MGMapNet: Multi-Granularity Representation Learning for End-to-End Vectorized HD Map Construction
74. ICLR 2025Navigation-Guided Sparse Scene Representation for End-to-End Autonomous Driving
75. ICLR 2025PETRA: Parallel End-to-end Training with Reversible Architectures
76. ICML 2025AutoCATE: End-to-End, Automated Treatment Effect Estimation
77. ICML 2025DSBRouter: End-to-end Global Routing via Diffusion Schr\"{o}dinger Bridge
78. ICML 2025End-to-End Learning Framework for Solving Non-Markovian Optimal Control
79. ICML 2025GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance
80. NeurIPS 2025Causal LLM Routing: End-to-End Regret Minimization from Observational Data
81. NeurIPS 2025DiffE2E: Rethinking End-to-End Driving with a Hybrid Diffusion-Regression-Classification Policy
82. NeurIPS 2025DriveDPO: Policy Learning via Safety DPO For End-to-End Autonomous Driving
83. NeurIPS 2025DrivingRecon: Large 4D Gaussian Reconstruction Model For Autonomous Driving
84. NeurIPS 2025Embodied Cognition Augmented End2End Autonomous Driving
85. NeurIPS 2025From Condensation to Rank Collapse: A Two-Stage Analysis of Transformer Training Dynamics
86. NeurIPS 2025Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution
87. NeurIPS 2025GaussianFusion: Gaussian-Based Multi-Sensor Fusion for End-to-End Autonomous Driving
88. NeurIPS 2025LinEAS: End-to-end Learning of Activation Steering with a Distributional Loss
89. NeurIPS 2025Overcoming Challenges of Long-Horizon Prediction in Driving World Models
90. NeurIPS 2025Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving
91. NeurIPS 2025Puzzles: Unbounded Video-Depth Augmentation for Scalable End-to-End 3D Reconstruction
92. NeurIPS 2025RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning
93. NeurIPS 2025Raw2Drive: Reinforcement Learning with Aligned World Models for End-to-End Autonomous Driving (in CARLA v2)
94. NeurIPS 2025Revisiting End-to-End Learning with Slide-level Supervision in Computational Pathology
95. NeurIPS 2025SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving
96. NeurIPS 2025SynCL: A Synergistic Training Strategy with Instance-Aware Contrastive Learning for End-to-End Multi-Camera 3D Tracking
97. NeurIPS 2025VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting
98. CVPR 2024DualAD: Disentangling the Dynamic and Static World for End-to-End Driving
99. CVPR 2024End-to-End Spatio-Temporal Action Localisation with Video Transformers
100. CVPR 2024End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames

常见问题

2023–2026 年四大会议一共有多少篇端到端驾驶论文？

共 161 篇，按年份分别是 2023 年 33 篇、2024 年 31 篇、2025 年 39 篇、2026 年 58 篇。

NeurIPS 2026 有多少篇端到端驾驶论文？

NeurIPS 2026 有 15 篇端到端驾驶论文，上一届是 18 篇。

ICLR 2026 有多少篇端到端驾驶论文？

ICLR 2026 有 12 篇端到端驾驶论文，上一届是 5 篇。

ICML 2026 有多少篇端到端驾驶论文？

ICML 2026 有 4 篇端到端驾驶论文，上一届是 4 篇。

CVPR 2026 有多少篇端到端驾驶论文？

CVPR 2026 有 27 篇端到端驾驶论文，上一届是 12 篇。

哪个会议的端到端驾驶论文最多？

CVPR，2023–2026 年共 71 篇。