Research map / Vision-language and multimodal / Robot action and navigation
One Scan is Enough: Demonstration-Free Adaptation for Language-Guided Navigation
Zhuoyuan Yu, Yuxing Long, Du Wang, Junyang Wang, Hongwei Fan, Hao Dong
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 Scan2Instruction, which turns a single brief iPhone scan of a building into a training environment and automatically synthesizes instructions to adapt a language-guided navigation policy without human demonstrations.
Problem
Language-guided navigation policies (VLAs) degrade in new buildings as layout, appearance, and language shift, and the standard fix of teleoperating new demonstrations on site does not scale.
Approach
Replaces human demonstrations with a single brief iPhone scan that is compiled into a complete training environment, from which navigation instructions and trajectories are automatically synthesized for on-site adaptation.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| post-train RealNav with PRPT-KS on Scan2Instruction episodes instead of generic pretraining only | RealNav · five scanned scenes, real robot | success rate: 0.25 -> 0.85 | Improved |
| use PRPT-KS on a single scan instead of fine-tuning on 20 human teleoperated trajectories per scene | RealNav · five scanned scenes, real robot | success rate: 0.49 -> 0.85 | Improved |
| use PRPT-KS adaptation versus human-demo SFT in never-scanned regions | RealNav · unscanned regions, real robot | success rate: 0.41 -> 0.82 | Improved |
| fine-tune on 20 human demonstrations per scene | RealNav and CorrectNav · five scanned scenes, real robot | success rate: 0.25 -> 0.49 (RealNav); 0.22 -> 0.48 (CorrectNav) | Improved |
| add generic navigation pretraining only (no scene adaptation) for RealNav | RealNav · Gaussian simulation, five scenes | success rate: 0.23 vs external baselines 0.20-0.23 | No clear effect |
| add the behavior-cloning anchor (planner-reference replay loss) (ablation) | RealNav + PRPT-KS · Gaussian simulation, five scenes | success rate: 0.30 -> 0.85 | Improved |
| add the Pareto-utility advantage policy-gradient term (ablation) | RealNav + PRPT-KS · Gaussian simulation, five scenes | success rate: 0.81 -> 0.85 | Improved |
| add the Pareto-weighted preference loss (ablation) | RealNav + PRPT-KS · Gaussian simulation, five scenes | success rate: 0.80 -> 0.85 | Improved |
| add the policy-gradient term or the preference loss (ablation) | RealNav + PRPT-KS · Gaussian simulation, five scenes | navigation error (m): 1.31 -> 1.42 (policy gradient); 1.32 -> 1.42 (preference) | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CorrectNav — CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model (AAAI 2026) · better on 2 of 2 shared results
- OmniNav — OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation (ICLR 2026) · better on 1 of 1 shared result
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
- General Scene Adaptation for Vision-and-Language Navigation (ICLR 2025 · similarity 0.90)
- A New Path: Scaling Vision-and-Language Navigation With Synthetic Instructions and Imitation Learning (CVPR 2023 · similarity 0.90)
- Test-Time Adaptation for Online Vision-Language Navigation with Feedback-based Reinforcement Learning (ICML 2025 · similarity 0.88)
- ImagineNav: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination (ICLR 2025 · similarity 0.87)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.87)
- STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization (NeurIPS 2025 · similarity 0.87)
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.
