AI Paper Atlas

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

FloVerse: Floor Plan-Guided Multi-Modal Navigation

Weiqi Huang, Shuangyi Dong, Jiaxin Li, Yifei Guo, Zan Wang, Wei Liang

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 1 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 floor plan input (ablation)ThreeDiff · PointNavSR: 25.8 -> 42.0Improved
add floor plan input (ablation)ThreeDiff · ImageNavSR: 22.6 -> 28.9Improved
add floor plan input (ablation)ThreeDiff · ObjectNavSR: 20.9 -> 28.6Improved
add floor plan input (ablation)ThreeDiff · ObjectNavSPL: 16.5 -> 22.3Improved
train jointly on three modalities with random masking vs Point-only (ablation)ThreeDiff · PointNavSPL: 38.9 -> 36.6Hurt
train jointly on three modalities vs Image-only (ablation)ThreeDiff · ImageNavSR: 25.4 -> 28.9Improved
train jointly on three modalities vs Object-only (ablation)ThreeDiff · ObjectNavSR: 24.3 -> 28.6Improved
add depth-based refiner (ablation)ThreeDiff · PointNavSR: 34.3 -> 42.0Improved
add depth-based refiner (ablation)ThreeDiff · ImageNavSR: 27.7 -> 28.9Improved
use ThreeDiff vs RL Skill Chain on GOAT ImageNavThreeDiff · GOAT-Bench ImageNavSR: 42.2 -> 31.0 (SR), 18.0 -> 25.6 (SPL)Mixed

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