Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2509.12894 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916952077762560 |
|---|---|
| author | Han, Leekyeung Min, Hyunji Hwangbo, Gyeom Choi, Jonghyun Seo, Paul Hongsuck |
| author_facet | Han, Leekyeung Min, Hyunji Hwangbo, Gyeom Choi, Jonghyun Seo, Paul Hongsuck |
| contents | We introduce DialNav, a novel collaborative embodied dialog task, where a navigation agent (Navigator) and a remote guide (Guide) engage in multi-turn dialog to reach a goal location. Unlike prior work, DialNav aims for holistic evaluation and requires the Guide to infer the Navigator's location, making communication essential for task success. To support this task, we collect and release the Remote Assistance in Navigation (RAIN) dataset, human-human dialog paired with navigation trajectories in photorealistic environments. We design a comprehensive benchmark to evaluate both navigation and dialog, and conduct extensive experiments analyzing the impact of different Navigator and Guide models. We highlight key challenges and publicly release the dataset, code, and evaluation framework to foster future research in embodied dialog. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_12894 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DialNav: Multi-turn Dialog Navigation with a Remote Guide Han, Leekyeung Min, Hyunji Hwangbo, Gyeom Choi, Jonghyun Seo, Paul Hongsuck Computer Vision and Pattern Recognition We introduce DialNav, a novel collaborative embodied dialog task, where a navigation agent (Navigator) and a remote guide (Guide) engage in multi-turn dialog to reach a goal location. Unlike prior work, DialNav aims for holistic evaluation and requires the Guide to infer the Navigator's location, making communication essential for task success. To support this task, we collect and release the Remote Assistance in Navigation (RAIN) dataset, human-human dialog paired with navigation trajectories in photorealistic environments. We design a comprehensive benchmark to evaluate both navigation and dialog, and conduct extensive experiments analyzing the impact of different Navigator and Guide models. We highlight key challenges and publicly release the dataset, code, and evaluation framework to foster future research in embodied dialog. |
| title | DialNav: Multi-turn Dialog Navigation with a Remote Guide |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.12894 |