AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

DriveFine: Refining-Augmented Masked Diffusion VLA for Accurate and Robust Driving

Chenxu Dang, Sining Ang, Yongkang Li, Haochen Tian, Jie Wang, Guang Li, Hangjun Ye, Jie Ma, Long Chen, Yan Wang

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GRPO reinforcement training to SFT (ablation)DriveFine (LLaDA) · NAVSIM-v1 navtestPDMS: 86.7 -> 89.6Improved
add refinement expert with offline RFT (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 89.6 -> 90.3Improved
add online RFT on top of offline RFT (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 90.3 -> 90.8Improved
add 1 refinement block (250M params) versus none (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 89.6 -> 90.0Improved
increase refinement blocks from 4 to 8 (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 90.8 -> 90.7No clear effect
increase GRPO group size from 2 to 8 (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 88.4 -> 89.6Improved
increase GRPO group size from 8 to 10 (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 89.6 -> 89.6No clear effect
add PDMS-oriented RFT to SFT model (ablation)DriveFine · NAVSIM-v1 navtestPDMS: 86.7 -> 90.5Improved
add PDMS-oriented RFT to SFT model (ablation)DriveFine · NAVSIM-v1 navtestEPDMS: 86.8 -> 89.1Improved
add PDMS-oriented RFT to SFT model (ablation)DriveFine · NAVSIM-v1 navtestLK: 97.9 -> 97.7No clear effect
add PDMS-oriented RFT to SFT model (ablation)DriveFine · NAVSIM-v1 navtestEC: 83.3 -> 83.8No clear effect
add score-based reinforcement fine-tuning with an extra scorerDriveFine · NAVSIM-v1 navtestPDMS: 90.7 -> 91.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy