AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Plug-and-Play Label Map Diffusion for Universal Goal-Oriented Navigation

Zhixuan Shen, Yijie Zeng, Shengxiang Luo, Tianrui Li, Haonan Luo

ICML 2026 · Image generation › Autoregressive and latent generation · approach: Retrieval and matching with generative features · 0 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 obstacle map network G_phi as structural prior (ablation)PLMD · ObjectNav HM3D_v0.2SR: 0.636 -> 0.665Improved
add obstacle map network G_phi as structural prior (ablation)PLMD · Instance-ImageNav HM3D_v0.2SR: 0.730 -> 0.776Improved
add obstacle map network G_phi as structural prior (ablation)PLMD · Multi-Robot ObjectNav HM3D_v0.2SR: 0.714 -> 0.762Improved
add obstacle maps to the label map (ablation)PLMD · ObjectNav HM3D_v0.2SR: 0.626 -> 0.665Improved
add obstacle maps to the label map (ablation)PLMD · Instance-ImageNav HM3D_v0.2SR: 0.727 -> 0.776Improved
add obstacle maps to the label map (ablation)PLMD · Multi-Robot ObjectNav HM3D_v0.2SR: 0.717 -> 0.762Improved
add HDBSCAN clustering for goal localization (ablation)PLMD · Instance-ImageNav HM3D_v0.2SR: 0.757 -> 0.776Improved
add HDBSCAN clustering for goal localization (ablation)PLMD · ObjectNav HM3D_v0.2SR: 0.657 -> 0.665Improved
use diffusion-generated maps instead of replacing visualized semantic map (ablation)PLMD · Multi-Robot ObjectNav HM3D_v0.2SR: 0.731 -> 0.762Improved
use SPADE fusion instead of CNN fusion (ablation)PLMD · Instance-ImageNav HM3D_v0.2SR: 0.728 -> 0.776Improved
use SPADE fusion instead of attention fusion (ablation)PLMD · Instance-ImageNav HM3D_v0.2SR: 0.716 -> 0.776Improved
increase diffusion steps from 25 to 100 (ablation)PLMD · Instance-ImageNav HM3D_v0.2 (Jetson)SR: 0.680 -> 0.747Improved

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