AI Paper Atlas

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

ColaVLA: Leveraging Cognitive Latent Reasoning for Hierarchical Parallel Trajectory Planning in Autonomous Driving

Qihang Peng, Xuesong Chen, Chenye Yang, Shaoshuai Shi, Hongsheng Li

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 13 citations (OpenAlex)

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 latent reasoning (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 32.2 -> 31.3Improved
add the rethink stage on top of reasoning (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.3 -> 30.4Improved
use Hierarchical Parallel Planner instead of MLP planner (ablation)ColaVLA · NeuroNCAPNeuroNCAP Score (avg): 1.05 -> 1.50Improved
use Hierarchical Parallel Planner instead of diffusion planner (ablation)ColaVLA · NeuroNCAPNeuroNCAP Score (avg): 1.02 -> 1.50Improved
use Hierarchical Parallel Planner instead of diffusion planner (ablation)ColaVLA · NeuroNCAP side collisionsNeuroNCAP Score (side): 1.43 -> 0.82Hurt
retain 256 tokens instead of 128 (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.2 -> 30.4Improved
retain 256 tokens instead of 320 (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.7 -> 30.4Improved
use Interpolate multi-scale strategy instead of single scale (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.9 -> 30.4Improved
use Interpolate instead of Sequential multi-scale (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.8 -> 30.4Improved
use Interpolate instead of Reverse multi-scale (ablation)ColaVLA · nuScenes ablationAvg L2 (cm): 31.4 -> 30.4Improved
use single scale instead of Interpolate for 1s horizon (ablation)ColaVLA · nuScenes ablationL2 1s (cm): 13.8 -> 14.0Hurt
do latent reasoning instead of text chain-of-thoughtColaVLA · Inference latencylatency (ms per frame): 3727 -> 727Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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