AI Paper Atlas

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

RAE-NWM: Navigation World Model in Dense Visual Representation Space

Mingkun Zhang, wangtian shen, Fan Zhang, Haijian Qin, Zihao Pei, Ziyang Meng

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 11 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
learned time-driven gate instead of simple addition (ablation)RAE-NWM · SACSoN planningATE / RPE: 3.34 -> 2.91 ATE, 0.79 -> 0.70 RPEImproved
learned time-driven gate instead of MLP fusion (ablation)RAE-NWM · SACSoN planningATE / RPE: 3.54 -> 2.91 ATE, 0.83 -> 0.70 RPEImproved
learned gate instead of scheduled gate (ablation)RAE-NWM · SACSoN planningATE / RPE: 3.31 -> 2.91 ATE, 0.78 -> 0.70 RPEImproved
flow matching generative objective instead of deterministic DINO-token regression (ablation)RAE-NWM · SACSoN planningATE / RPE: 3.14 -> 2.91 ATE, 0.75 -> 0.70 RPEImproved
flow matching instead of deterministic DINO-token regression (ablation)RAE-NWM · RECON planningRPE: 0.36 vs. 0.37No clear effect
DINOv2 dense representation space instead of SD-VAE latent (ablation)RAE-NWM · long-horizon rolloutLPIPSMixed
add DDT head (ablation)RAE-NWM · long-horizon rolloutLPIPSImproved
RAE-NWM instead of NWM on direct 16s predictionRAE-NWM vs NWM · SACSoN direct prediction 16sFID: 33.06 -> 15.90Improved
RAE-NWM instead of NWM on direct 16s predictionRAE-NWM vs NWM · SACSoN direct prediction 16sLPIPS: 0.470 -> 0.349Improved
RAE-NWM instead of NWM in planning on RECONRAE-NWM vs NWM · RECON planning (2s)ATE: 1.13 -> 1.36Hurt
RAE-NWM instead of NWM in planning on SACSoNRAE-NWM vs NWM · SACSoN planning (2s)ATE: 4.12 -> 2.91Improved
RAE-NWM in Habitat versus One-Step WMRAE-NWM vs One-Step WM · Habitat image-goal navigationSR / SPL: SR 72.67 -> 78.95, SPL 69.10 -> 63.58Mixed

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