paper-with-me

Papers

Poutine: Vision-Language-Trajectory Pre-Training and Reinforcement Learning Post-Training Enable Robust End-to-End Autonomous Driving

2025-06-12 · Luke Rowe, Rodrigue de Schaetzen, Roger Girgis, Christopher Pal, Liam Paull

We present Poutine, a 3B-parameter vision-language model (VLM) tailored for end-to-end autonomous driving in long-tail driving scenarios. Poutine is trained in two stages. To obtain strong base driving capabilities, we train Poutine-Base in a self-supervised vision-language-trajectory (VLT) next-token prediction fashion on 83 hours of CoVLA nominal driving and 11 hours of Waymo long-tail driving. Accompanying language annotations are auto-generated with a 72B-parameter VLM. Poutine is obtained by fine-tuning Poutine-Base with Group Relative Policy Optimization (GRPO) using less than 500 preference-labeled frames from the Waymo validation set. We show that both VLT pretraining and RL fine-tuning are critical to attain strong driving performance in the long-tail. Poutine-Base achieves a rater-feedback score (RFS) of 8.12 on the validation set, nearly matching Waymo's expert ground-truth RFS. The final Poutine model achieves an RFS of 7.99 on the official Waymo test set, placing 1st in the 2025 Waymo Vision-Based End-to-End Driving Challenge by a significant margin. These results highlight the promise of scalable VLT pre-training and lightweight RL fine-tuning to enable robust and generalizable autonomy.

📄 PDF Abstract BibTeX arXiv:2506.11234

Code (0)

등록된 구현이 없습니다.

Tasks

Autonomous Driving

Methods 이 논문이 사용한 방법론

BASE 설명 없음

Similar Papers 제목 키워드 기반

Pyro: Deep Universal Probabilistic Programming

2018-10-18 · Eli Bingham, Jonathan P. Chen, Martin Jankowiak, Fritz Obermeyer 외

Pyro is a probabilistic programming language built on Python as a platform for developing advanced probabilistic models in AI research. To scale to large datasets and high-dimensional models, Pyro uses stochastic variati…

GPUProbabilistic ProgrammingVariational Inference

Is Your Trajectory Displacement Safe in Long-tail?

2026-06-15 · Qiao Sun, Weicheng Zheng, Yixin Huang, Hang Zhao arxiv

Long-tail scenarios remain a major bottleneck for autonomous driving evaluation, even as datasets grow by orders of magnitude. Existing evaluation pipelines are rarely human-aligned, safety-aware, verifiable, and explain…

Autonomous Driving

Amortizing intractable inference in diffusion models for vision, language, and control

2024-05-31 · Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim 외

Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors in downstream tasks poses an intractable posterior inference problem. This paper…

continuous-controlContinuous ControlDeep Reinforcement LearningImage Generation+4

Temporal GRPO: Beyond Trajectory-Level Credit in Vision-Language-Action Reinforcement Learning

2026-08-13 · Yao Zhou, Hang Gao, Fengge Wu, Changwen Zheng 외 arxiv

Outcome-driven reinforcement learning offers a scalable way to post-train vision-language-action (VLA) policies from sparse task-success feedback. In common GRPO-based VLA post-training, one rollout-level advantage is ap…

Reinforcement Learning

STARE-VLA: Progressive Stage-Aware Reinforcement for Fine-Tuning Vision-Language-Action Models

2025-12-04 · Feng Xu, Guangyao Zhai, Xin Kong, Tingzhong Fu 외 arxiv

Recent advances in Vision-Language-Action (VLA) models, powered by large language models and reinforcement learning-based fine-tuning, have shown remarkable progress in robotic manipulation. Existing methods often treat …

Reinforcement Learning