AI Paper Atlas

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

Think Densely, Act Sparsely: Latent Expert Cognitive Chains for Vision-Language-Action Autonomous Driving

Jie Wang, Guang Li, Zhijian Huang, Jinlong Li, Chenxu Dang, Hangjun Ye, Yahong Han, Long Chen

NeurIPS 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes latent expert cognitive chains for vision-language-action autonomous driving, aiming to think densely in latent space while acting with sparse trajectory outputs.

Problem

Existing VLA driving models are trained with sparse supervision (language instructions or trajectories), forcing compression or loss of dense visual, semantic, and geometric scene information needed for driving.

Approach

Introduces latent expert cognitive chains that let the model reason densely over visual, semantic, and geometric information in latent form before producing sparse driving actions, instead of mapping dense observations directly to sparse outputs.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the full latent expert cognitive chain (DINOv3, SAM 3, DepthAnything 3) with hybrid trajectory formulation (ablation)Qwen2.5-VL-7B baseline · NAVSIM-v1PDMS: 86.04 -> 89.81Improved
replace direct coordinate regression with discrete motion primitive retrieval (ablation)Qwen2.5-VL-7B baseline · NAVSIM-v1PDMS: 86.04 -> 87.44Improved
add continuous offset refinement on top of motion primitives (ablation)motion primitive trajectory head · NAVSIM-v1PDMS: 87.44 -> 88.49Improved
add DINOv3 expert latent tokens (ablation)MT + Offset · NAVSIM-v1PDMS: 88.49 -> 89.07Improved
add SAM 3 semantic expert after DINOv3 (ablation)MT + Offset + DINOv3 · NAVSIM-v1PDMS: 89.07 -> 89.31Improved
add DepthAnything 3 geometric expert after DINOv3 and SAM 3 (ablation)MT + Offset + DINOv3 + SAM 3 · NAVSIM-v1PDMS: 89.31 -> 89.81Improved
add task-specific latent tokens instead of using visual tokens with task heads (ablation)LECDrive · NAVSIM-v1PDMS: 88.73 -> 89.81Improved
aggregate latent embeddings from four VLM layers instead of the final layer only (ablation)LECDrive · NAVSIM-v1PDMS: 89.26 -> 89.81Improved
use 8 latent tokens per expert instead of 2 (ablation)LECDrive · NAVSIM-v1PDMS: 89.13 -> 89.81Improved
use 7B instead of 3B base VLM (ablation)LECDrive · NAVSIM-v1PDMS: 89.21 -> 89.81Improved
add all three experts without motion-primitive/offset trajectory formulation (ablation)Qwen2.5-VL-7B baseline · NAVSIM-v1PDMS: 86.04 -> 89.15Improved
change expert order from DINO-SAM-DEPTH to DINO-DEPTH-SAM or SAM-DINO-DEPTH (ablation)LECDrive · NAVSIM-v1PDMS: 89.81 -> 89.18 and 89.66Hurt

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