Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Yue, Ma, Tianyi, Wang, Zun, Qiao, Yanyuan, Kordjamshidi, Parisa
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918150780485632
author Zhang, Yue
Ma, Tianyi
Wang, Zun
Qiao, Yanyuan
Kordjamshidi, Parisa
author_facet Zhang, Yue
Ma, Tianyi
Wang, Zun
Qiao, Yanyuan
Kordjamshidi, Parisa
contents Integrating large language models (LLMs) into embodied AI models is becoming increasingly prevalent. However, existing zero-shot LLM-based Vision-and-Language Navigation (VLN) agents either encode images as textual scene descriptions, potentially oversimplifying visual details, or process raw image inputs, which can fail to capture abstract semantics required for high-level reasoning. In this paper, we improve the navigation agent's contextual understanding by incorporating textual descriptions from multiple perspectives that facilitate analogical reasoning across images. By leveraging text-based analogical reasoning, the agent enhances its global scene understanding and spatial reasoning, leading to more accurate action decisions. We evaluate our approach on the R2R dataset, where our experiments demonstrate significant improvements in navigation performance.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25139
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs
Zhang, Yue
Ma, Tianyi
Wang, Zun
Qiao, Yanyuan
Kordjamshidi, Parisa
Artificial Intelligence
Computer Vision and Pattern Recognition
Multimedia
Integrating large language models (LLMs) into embodied AI models is becoming increasingly prevalent. However, existing zero-shot LLM-based Vision-and-Language Navigation (VLN) agents either encode images as textual scene descriptions, potentially oversimplifying visual details, or process raw image inputs, which can fail to capture abstract semantics required for high-level reasoning. In this paper, we improve the navigation agent's contextual understanding by incorporating textual descriptions from multiple perspectives that facilitate analogical reasoning across images. By leveraging text-based analogical reasoning, the agent enhances its global scene understanding and spatial reasoning, leading to more accurate action decisions. We evaluate our approach on the R2R dataset, where our experiments demonstrate significant improvements in navigation performance.
title Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs
topic Artificial Intelligence
Computer Vision and Pattern Recognition
Multimedia
url https://arxiv.org/abs/2509.25139