AI Paper Atlas

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

Reasoning-VLA: An Efficient and Spatial-Guided General Vision-Language-Action Reasoning Model for Autonomous Driving

Dapeng Zhang, Zhenlong Yuan, Zhangquan Chen, Chih-Ting Liao, Yinda Chen, Fei Shen, Qingguo Zhou, Tat-Seng Chua

ICML 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 learnable action queries instead of non-learnable queries (ablation)Reasoning-VLA-7B · nuScenes subset of unified datasetavg L2 (m): 0.30 -> 0.23Improved
add implicit spatial guidance initialization of action queries (ablation)Reasoning-VLA-7B · nuScenes subset of unified datasetavg L2 (m): 0.27 -> 0.23Improved
add action refinement module (ablation)Reasoning-VLA-7B · nuScenes subset of unified datasetavg L2 (m): 0.29 -> 0.23Improved
add RL (GRPO) fine-tuning after SFT (ablation)Reasoning-VLA-7B · nuScenes subset of unified datasetavg L2 (m): 0.26 -> 0.23Improved
add RL after SFT when ARM is removed (ablation)R-VLA(w/o ARM) · nuScenes subset of unified datasetavg L2 (m): 0.29 -> 0.29No clear effect
add RL after SFT on zero-shot unseen datasets (ablation)Reasoning-VLA-7B · Zero-shot unified datasetavg L2 (m): 0.31 -> 0.29Improved
add RL to SFT on the unified dataset (per sub-dataset) (ablation)Reasoning-VLA-7B · Unified dataset sub-datasetsavg L2 (m): 0.24 -> 0.23No clear effect
scale backbone from 3B to 7BReasoning-VLA · nuScenes open-loopavg L2 (m): 0.30 -> 0.23Improved
additional RL on nuScenes training clips (7B+)Reasoning-VLA-7B · nuScenes open-loopavg L2 (m): 0.23 -> 0.22Improved
additional RL on nuScenes training clips (7B+)Reasoning-VLA-7B · NeuroNCAP closed-loopavg NeuroNCAP score: 2.25 -> 2.19Hurt
additional RL on nuScenes training clips (7B+)Reasoning-VLA-7B · NeuroNCAP closed-loopavg collision rate (%): 59.4 -> 59.8Hurt
use Reasoning-VLA-7B instead of Para-Drive on NAVSIMNAVSIM planner · NAVSIMPDMS: 84.0 -> 91.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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