This work turns demonstration-free scene adaptation for language-guided navigation from an aspiration into a working reality, and replaces human demonstration with a single brief iPhone scan.
Abstract
Deploying a language-guided navigation policy in a new building usually means watching it fail. Public-benchmark VLAs degrade scene by scene as layout, appearance, and language drift, and the standard fix of teleoperating fresh trajectories on site does not scale. We replace human demonstration with a single brief iPhone scan. Scan2Instruction compiles that scan into a complete training environment, automatically synthesizing executable navigation episodes with grounded instructions. PRPT-KS then adapts the policy through a closed-form Pareto utility read directly off the reconstructed scene, with no learned critic and no human labels. Across five real environments, the two designs together drive our RealNav to an 85% real-robot success rate, compared with 49% for a RealNav baseline post-trained on human demonstrations, and still achieve 82% in unscanned regions of the same buildings, compared with 41% for the same baseline. Our work turns demonstration-free scene adaptation for language-guided navigation from an aspiration into a working reality.
FINE is introduced, a Future-Informed Navigation Encoding framework that extracts this latent supervision from existing demonstrations and improves success rate on R2R-CE and RxR-CE val-unseen, recovering roughly one-third of the performance lost by reducing the training demonstrations.
Khang Nguyen, Hoang Pham Quang Nguyen, Ha Phuong Nguyen et al.· arXiv.org· 0 citations
Vision-and-Language Navigation (VLN) requires embodied agents to ground language in visual observations and make long-horizon decisions. Existing methods often address these demands through task-specific architectures. As multimodal large language models (MLLMs) become increasingly capable, an important question remain...
Ling-Yi Hong, Wei-Ren Cai, Jia-Yu Chen et al.· 1 citation
Recent navigation systems achieve strong benchmark results, yet real-world deployment often remains visibly stop-and-go. This bottleneck arises because the sense-inference-execution loop is still blocking: after each new observation, the controller must wait for sensing, transmission, and inference before motion can co...
Xiang-Chen Wang, Wei-Ye Zhu, Teng Wang et al.· IEEE Robotics and Automation...· 0 citations
VCN-Bench is introduced, a VCN-Bench benchmark for probing closed-loop spatial reasoning over prior visual experience in MLLMs, and MV-DualVLN is proposed, a planning-oriented baseline that jointly leverages prior video and in-episode observations.
Si-Qi Zhang, Meng Wei, Chen-Yang Wan et al.· arXiv.org· 0 citations
Aerial vision-language navigation (VLN) agents built on large vision-language models are expensive to query. Executing a complete predicted action chunk before replanning amortizes this cost, reducing model inference from 195 to 42 ms per executed waypoint for our policy. It also moves the states at which the policy is...
Vision-and-language navigation (VLN) relies on retaining and using relevant observations. Yet growing histories introduce redundancy, and retained information may remain underused. We introduce ReInsBench, a diagnostic benchmark that examines history utilization through controlled continuation trials. Its results show...