Pocket RAG: On-Device RAG for First Aid Guidance in Offline Mobile Environment
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910021922586624 |
|---|---|
| author | Kang, Dong Ho Lee, Hyunjoon Cha, Hyeonjeong Choi, Minkyu Lim, Sungsoo |
| author_facet | Kang, Dong Ho Lee, Hyunjoon Cha, Hyeonjeong Choi, Minkyu Lim, Sungsoo |
| contents | In disaster scenarios or remote areas, first responders often lose network connectivity when providing first aid. In such situations, server-based AI systems fail to provide critical guidance. To address this issue, we present a lightweight, mobile-based retrieval-augmented generation system for small language models (SLMs) that can run directly on Android devices. Our system integrates a mobile-friendly optimized pipeline featuring Hybrid RAG, selective compression, batched prompt decoding, and quantization caching. Despite the model's small size, our RAG-based system achieves 94.5\% accuracy for physical first aid and 97.0\% for psychological first aid. Additionally, we reduce response time from 14.2s to 3.7s, achieving a nearly 4x speedup. These results prove that our system is practical and can deliver reliable first aid guidance even without internet connectivity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13229 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Pocket RAG: On-Device RAG for First Aid Guidance in Offline Mobile Environment Kang, Dong Ho Lee, Hyunjoon Cha, Hyeonjeong Choi, Minkyu Lim, Sungsoo Networking and Internet Architecture In disaster scenarios or remote areas, first responders often lose network connectivity when providing first aid. In such situations, server-based AI systems fail to provide critical guidance. To address this issue, we present a lightweight, mobile-based retrieval-augmented generation system for small language models (SLMs) that can run directly on Android devices. Our system integrates a mobile-friendly optimized pipeline featuring Hybrid RAG, selective compression, batched prompt decoding, and quantization caching. Despite the model's small size, our RAG-based system achieves 94.5\% accuracy for physical first aid and 97.0\% for psychological first aid. Additionally, we reduce response time from 14.2s to 3.7s, achieving a nearly 4x speedup. These results prove that our system is practical and can deliver reliable first aid guidance even without internet connectivity. |
| title | Pocket RAG: On-Device RAG for First Aid Guidance in Offline Mobile Environment |
| topic | Networking and Internet Architecture |
| url | https://arxiv.org/abs/2602.13229 |