mlsanityData to 14 Oct 2026

FocusDrive: Reasoning with Visual Focus for Autonomous Driving

arXiv 2609.33190 ↗First posted 27 Sep 2026cs.CV cs.ROcited by 0

On the map: End-to-end autonomous driving → VLMs for autonomous driving also fits: Reasoning VLA and VLM driving policies

Abstract (arXiv)

Driving decisions depend on both where to focus and how to act on what is seen. Effective driving reasoning must establish which objects matter, where they are, and how they inform the intended action. Text-based rationales can describe a driving response while leaving its correspondence to specific visual evidence implicit. Visual focus provides a concrete starting point for this connection by identifying what matters in the scene and where it is. We propose FocusDrive, a structured multimodal reasoning framework that organizes end-to-end planning around explicit visual focus. It pairs descriptions of decision-relevant objects with image-patch references, bringing explicit visual focus into the reasoning that generates driving plans and trajectories. We first assess this focus representation through driver gaze prediction, then investigate its role in planning reasoning using driving-focus annotations within existing NAVSIM training scenes. Experiments on W3DA and NAVSIM demonstrate competitive gaze prediction and end-to-end planning performance, with FocusDrive improving over text-based chain-of-thought. These results support visual focus as an effective link between scene understanding and driving action.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in VLMs for autonomous driving, one of 9 directions in End-to-end autonomous driving. Builds on 35 of its 42 references that are in the map. First posted on arXiv 27 Sep 2026.

Builds on

35 of 42 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025FSDAM: Few-Shot Driving Attention Modeling via Vision-Language Coupling16 Nov
  2. 2024DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models19 Feb
  3. 2023DriveLM: Driving with Graph Visual Question Answering21 Dec

Other references · 32

  1. 2022A ConvNet for the 2020scited by 3,579
  2. 2022TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Drivingcited by 485
  3. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Modelscited by 451
  4. 2024DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Drivingcited by 334
  5. 2025AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuningcited by 260
  6. 2025ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Drivingcited by 198
  7. 2025ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generationcited by 178
  8. 2025FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Drivingcited by 166
  9. 2025End-to-End Driving with Online Trajectory Evaluation via BEV World Modelcited by 151
  10. 2025DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Drivingcited by 134
  11. 2017Predicting the Driver's Focus of Attention: the DR(eye)VE Projectcited by 79
  12. 2025Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokenscited by 71
20 more
  1. 2019DADA: Driver Attention Prediction in Driving Accident Scenarioscited by 70
  2. 2017Predicting Driver Attention in Critical Situationscited by 61
  3. 2025AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Drivingcited by 58
  4. 2019DADA-2000: Can Driving Accident be Predicted by Driver Attention? Analyzed by A Benchmarkcited by 57
  5. 2016A Deep Multi-Level Network for Saliency Predictioncited by 54
  6. 2024Making Large Language Models Better Planners with Reasoning-Decision Alignmentcited by 52
  7. 2025DriveDPO: Policy Learning via Safety DPO For End-to-End Autonomous Drivingcited by 39
  8. 2026DriveWorld-VLA: Unified Latent-Space World Modeling with Vision-Language-Action for Autonomous Drivingcited by 37
  9. 2025Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMscited by 20
  10. 2026DriveFuture: Future-Aware Latent World Models for Autonomous Drivingcited by 17
  11. 2024GazeXplain: Learning to Predict Natural Language Explanations of Visual Scanpathscited by 15
  12. 2026Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Modelscited by 15
  13. 2026AutoDrive-P3: Unified Chain of Perception-Prediction-Planning Thought via Reinforcement Fine-Tuningcited by 14
  14. 2026DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Drivingcited by 11
  15. 2025Where, What, Why: Towards Explainable Driver Attention Predictioncited by 10
  16. 2017Visual Saliency Prediction Using a Mixture of Deep Neural Networkscited by 7
  17. 2019Human Visual Attention Prediction Boosts Learning & Performance of Autonomous Driving Agentscited by 7
  18. 2026Towards Pixel-Level VLM Perception via Simple Points Predictioncited by 5
  19. 2026Driving risk emerges from the required two-dimensional joint evasive accelerationcited by 1
  20. 2026From Scene to Object: Text-Guided Dual-Gaze Predictioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025InsightDrive: Insight Scene Representation for End-to-End Autonomous Driving
  2. 2025DriveBLIP2: Attention-Guided Explanation Generation for Complex Driving Scenarios
  3. 2026MindDriver: Introducing Progressive Multimodal Reasoning for Autonomous Driving
  4. 2026FactorDrive: Adaptive Multi-Step Reasoning Driven by Planning-Critical Factors for End-to-End Autonomous Driving
  5. 2026SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving
  6. 2025DriveRX: A Vision-Language Reasoning Model for Cross-Task Autonomous Driving
  7. 2026Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving
  8. 2026Drive by Hindsight and Foresight: Tool-Grounded Synergistic Reasoning over Hierarchical Memory for Autonomous Driving