mlsanityData to 14 Oct 2026

Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail

arXiv 2511.00088 ↗First posted 30 Oct 2025cs.RO cs.AI cs.LGcited by 146

On the map: LLM reasoning → Chain-of-thought reasoning for driving

also in 3 more
Abstract (arXiv)

End-to-end architectures trained via imitation learning have advanced autonomous driving by scaling model size and data, yet performance remains brittle in safety-critical long-tail scenarios where supervision is sparse and causal understanding is limited. We introduce Alpamayo-R1 (AR1), a vision-language-action model (VLA) that integrates Chain of Causation reasoning with trajectory planning for complex driving scenarios. Our approach features three key innovations: (1) the Chain of Causation (CoC) dataset, built through a hybrid auto-labeling and human-in-the-loop pipeline producing decision-grounded, causally linked reasoning traces aligned with driving behaviors; (2) a modular VLA architecture combining Cosmos-Reason, a vision-language model pre-trained for Physical AI, with a diffusion-based trajectory decoder that generates dynamically feasible trajectories in real time; (3) a multi-stage training strategy using supervised fine-tuning to elicit reasoning and reinforcement learning (RL) to enforce reasoning-action consistency and optimize reasoning quality. AR1 achieves up to a 12% improvement in planning accuracy on challenging cases compared to a trajectory-only baseline, with a 35% reduction in close encounter rate in closed-loop simulation. RL post-training improves reasoning quality by 45% and reasoning-action consistency by 37%. Model scaling from 0.5B to 7B parameters shows consistent improvements. On-vehicle road tests confirm real-time performance (99 ms latency) and successful urban deployment. By bridging interpretable reasoning with precise control, AR1 demonstrates a practical path towards Level 4 autonomous driving. Model weights are available at https://huggingface.co/nvidia/Alpamayo-R1-10B with inference code at https://github.com/NVlabs/alpamayo.

Key facts. Cited by 146 papers in the map (data through 14 Oct 2026). Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning. Builds on 89 of its 118 references that are in the map. First posted on arXiv 30 Oct 2025.

Builds on

89 of 118 references.

Closest ancestors · 15

Same direction, newest first

  1. 2025MTRDrive: Memory-Tool Synergistic Reasoning for Robust Autonomous Driving in Corner Cases25 Sep
  2. 2025AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving17 Sep
  3. 2025ReAL-AD: Towards Human-Like Reasoning in End-to-End Autonomous Driving16 Jul
  4. 2025Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning23 Jun
  5. 2025AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning16 Jun
  6. 2025ReasonPlan: Unified Scene Prediction and Decision Reasoning for Closed-loop Autonomous Driving26 May
  7. 2025AgentThink: A Unified Framework for Tool-Augmented Chain-of-Thought Reasoning in Vision-Language Models for Autonomous Driving21 May
  8. 2025DSDrive: Distilling Large Language Model for Lightweight End-to-End Autonomous Driving with Unified Reasoning and Planning8 May
  9. 2025DriveAgent: Multi-Agent Structured Reasoning with LLM and Multimodal Sensor Fusion for Autonomous Driving4 May
  10. 2025NuScenes-SpatialQA: A Spatial Understanding and Reasoning Benchmark for Vision-Language Models in Autonomous Driving4 Apr
  11. 2025CoT-Drive: Efficient Motion Forecasting for Autonomous Driving with LLMs and Chain-of-Thought Prompting10 Mar
  12. 2024VLM-AD: End-to-End Autonomous Driving through Vision-Language Model Supervision19 Dec
3 more
  1. 2024WOMD-Reasoning: A Large-Scale Dataset for Interaction Reasoning in Driving5 Jul
  2. 2024DriveCoT: Integrating Chain-of-Thought Reasoning with End-to-End Driving25 Mar
  3. 2023Reason2Drive: Towards Interpretable and Chain-based Reasoning for Autonomous Driving6 Dec

Other references · 74

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  5. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  6. 2020Scaling Laws for Neural Language Modelscited by 6,586
  7. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  8. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  9. 2022Flow Matching for Generative Modelingcited by 5,702
  10. 2025Qwen2.5-VL Technical Reportcited by 5,587
  11. 2017Neural Discrete Representation Learningcited by 4,712
  12. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
38 more
  1. 2017Deep reinforcement learning from human preferencescited by 3,912
  2. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  3. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  4. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  5. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  6. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  7. 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
  8. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,046
  9. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,761
  10. 2025SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Featurescited by 1,188
  11. 2022Planning-oriented Autonomous Drivingcited by 1,087
  12. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,061
  13. 2016End to End Learning for Self-Driving Carscited by 1,009
  14. 2023RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedbackcited by 625
  15. 2023VAD: Vectorized Scene Representation for Efficient Autonomous Drivingcited by 622
  16. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 537
  17. 2023DriveLM: Driving with Graph Visual Question Answeringcited by 536
  18. 2023DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Modelcited by 461
  19. 2023Aligning Text-to-Image Models using Human Feedbackcited by 422
  20. 2023GPT-Driver: Learning to Drive with GPTcited by 347
  21. 2023NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenariocited by 262
  22. 2016A Survey of Motion Planning and Control Techniques for Self-driving Urban Vehiclescited by 254
  23. 2024Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Drivingcited by 252
  24. 2023Parting with Misconceptions about Learning-based Vehicle Motion Planningcited by 242
  25. 2018Textual Explanations for Self-Driving Vehiclescited by 226
  26. 2023MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusioncited by 213
  27. 2025ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Drivingcited by 198
  28. 2022Guided Conditional Diffusion for Controllable Traffic Simulationcited by 190
  29. 2025OpenDriveVLA: Towards End-to-end Autonomous Driving with Large Vision Language Action Modelcited by 176
  30. 2025FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Drivingcited by 166
  31. 2023A Language Agent for Autonomous Drivingcited by 164
  32. 2024Senna: Bridging Large Vision-Language Models and End-to-End Autonomous Drivingcited by 156
  33. 2025SimLingo: Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignmentcited by 150
  34. 2025Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Bettercited by 146
  35. 2023LingoQA: Visual Question Answering for Autonomous Drivingcited by 145
  36. 2022Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarioscited by 129
  37. 2022DRAMA: Joint Risk Localization and Captioning in Drivingcited by 126
  38. 2025Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectivescited by 122

50 most cited of 74 All 74 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 11; 2026: 135.

146 papers.

Sort:

By area:

  1. 2026Driving on Registerscited by 44
  2. 2025Vision-Language-Action Models for Autonomous Driving: Past, Present, and Futurecited by 33
  3. 2026LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Drivingcited by 28
  4. 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planningcited by 24
  5. 2025Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoningcited by 22
  6. 2026UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Drivingcited by 19
  7. 2026SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Modelcited by 17
  8. 2025WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Drivingcited by 16
  9. 2025Latent Chain-of-Thought World Modeling for End-to-End Drivingcited by 15
  10. 2026NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoningcited by 15
  11. 2025dVLM-AD: Enhance Diffusion Vision-Language-Model for Driving via Controllable Reasoningcited by 14
  12. 2026Agentic World Modeling: Foundations, Capabilities, Laws, and Beyondcited by 14
38 more
  1. 2026DriveWAM: Video Generative Priors Enable Scalable World-Action Modeling for Autonomous Drivingcited by 14
  2. 2026NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulationcited by 14
  3. 2026ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Drivingcited by 12
  4. 2026Unleashing the Potential of Diffusion Models for End-to-End Autonomous Drivingcited by 11
  5. 2026Is the Future Compatible? Diagnosing Dynamic Consistency in World Action Modelscited by 11
  6. 2026Metis: A Generalizable and Efficient World-Action Model for Autonomous Driving and Urban Navigationcited by 11
  7. 2025UniUGP: Unifying Understanding, Generation, and Planing For End-to-end Autonomous Drivingcited by 10
  8. 2026Xiaomi OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanationcited by 10
  9. 2025Percept-WAM: Perception-Enhanced World-Awareness-Action Model for Robust End-to-End Autonomous Drivingcited by 9
  10. 2026AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture-of-Transformers for End-to-End Autonomous Drivingcited by 9
  11. 2026BridgeSim: Unveiling the OL-CL Gap in End-to-End Autonomous Drivingcited by 9
  12. 2026nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Drivingcited by 9
  13. 2026SimWAM: A Simple World Action Model for End-to-End Autonomous Drivingcited by 9
  14. 2026Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Drivingcited by 9
  15. 2026Learning from Mistakes: Post-Training for Driving VLA with Takeover Datacited by 8
  16. 2026What-If World: A Causal Benchmark for General World Models in Embodied Scenarioscited by 8
  17. 2025RoaD: Rollouts as Demonstrations for Closed-Loop Supervised Fine-Tuning of Autonomous Driving Policiescited by 6
  18. 2026Accelerating Structured Chain-of-Thought in Autonomous Vehiclescited by 6
  19. 2026StyleVLA: Driving Style-Aware Vision Language Action Model for Autonomous Drivingcited by 6
  20. 2026MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Drivingcited by 6
  21. 2026HERMES: A Holistic End-to-End Risk-Aware Multimodal Embodied System with Vision-Language Models for Long-Tail Autonomous Drivingcited by 5
  22. 2026Action Hallucination in Generative Vision-Language-Action Modelscited by 5
  23. 2026NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Drivingcited by 4
  24. 2026Reasoning models do not yet follow their reasoning in autonomous driving: The KITScenes LongTail Datasetcited by 4
  25. 2026Scaling Self-Play for End-to-End Drivingcited by 4
  26. 2025Foundation Models for Trajectory Planning in Autonomous Driving: A Review of Progress and Open Challengescited by 3
  27. 2026Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoningcited by 3
  28. 2026Latent World Models for Automated Driving: A Unified Taxonomy, Evaluation Framework, and Open Challengescited by 3
  29. 2026ADV-0: Closed-Loop Min-Max Adversarial Training for Long-Tail Robustness in Autonomous Drivingcited by 3
  30. 2026The Era of End-to-End Autonomy: Transitioning from Rule-Based Driving to Large Driving Modelscited by 3
  31. 2026123D: Unifying Multi-Modal Autonomous Driving Data at Scalecited by 3
  32. 2026Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Drivingcited by 3
  33. 2026Lost in Fog: Sensor Perturbations Expose Reasoning Fragility in Driving VLAscited by 3
  34. 2026Planning-aligned Token Compression for Long-Context Autonomous Drivingcited by 3
  35. 2026VLGA: Vision-Language-Geometry-Action Models for Autonomous Drivingcited by 3
  36. 2026M*: A Modular, Extensible, Serving System for Multimodal Modelscited by 3
  37. 2026Instant NuRec: Feed-Forward 3D Gaussian Reconstruction for Driving Scene Simulationcited by 3
  38. 2026GeoWAM: Visual Geometry World Action Models for Autonomous Drivingcited by 3

50 most cited of 146 All 146 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025ColaVLA: Leveraging Cognitive Latent Reasoning for Hierarchical Parallel Trajectory Planning in Autonomous Driving
  2. 2025Reasoning-VLA: A Fast and General Vision-Language-Action Reasoning Model for Autonomous Driving
  3. 2026DriveMA: Rethinking Language Interfaces in Driving VLAs with One-Step Meta-Actions
  4. 2025DriveAgent-R1: Advancing VLM-based Autonomous Driving with Active Perception and Hybrid Thinking
  5. 2025MindDrive: A Vision-Language-Action Model for Autonomous Driving via Online Reinforcement Learning
  6. 2025DriveRX: A Vision-Language Reasoning Model for Cross-Task Autonomous Driving
  7. 2025CoT4AD: A Vision-Language-Action Model with Explicit Chain-of-Thought Reasoning for Autonomous Driving
  8. 2026CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving