Abstract

Traditional CRM knowledge systems remain heavily dependent on cloud processing, where latency, data privacy, and network availability pose significant challenges. The author presents an Edge-First Retrieval-Augmented Generation system for mobile CRM applications. It runs every information-retrieval and text-generation task on the user's mobile device or an edge server nearby, without allowing sensitive customer data to leave the device. To implement the prototype, a lightweight, on-device generative text and semantic search process is used, executed locally. The system has been tested with a custom-built synthetic dataset called 'CRM-410', which includes 410 anonymized customer interaction profiles. This has been performed primarily to measure and compare a quantified edge-first system against a traditional cloud-based baseline across three key axes: query-to-response time (latency), data exfiltration risk or privacy, and functionality during network loss or unavailability. These results demonstrate that edge-first cuts latency to less than 2 seconds, enforces complete data privacy by keeping information local, and provides a strong, viable alternative for responsive, secure mobile CRM professionals.

References

• Lewis, Patrick, Ethan Perez, Aleksandara Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Kuttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel and Douwe Kiela. “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.” ArXiv abs/2005.11401 (2020): n. pag.
• Guu, Kelvin, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang. “REALM: Retrieval-Augmented Language Model Pre-Training.” ArXiv abs/2002.08909 (2020): n. pag.
• Karpukhin, Vladimir, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Yu Wu, Sergey Edunov, Danqi Chen and Wen-tau Yih. “Dense Passage Retrieval for Open-Domain Question Answering.” ArXiv abs/2004.04906 (2020): n. pag.
• Gao, Yunfan, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Qianyu Guo, Meng Wang and Haofen Wang. “Retrieval-Augmented Generation for Large Language Models: A Survey.” ArXiv abs/2312.10997 (2023): n. pag.
• Edge, Darren, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva N. Mody, Steven Truitt and Jonathan Larson. “From Local to Global: A Graph RAG Approach to Query-Focused Summarization.” ArXiv abs/2404.16130 (2024): n. pag.
• Sarthi, Parth, Salman Abdullah, Aditi Tuli, Shubh Khanna, Anna Goldie and Christopher D. Manning. “RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval.” ArXiv abs/2401.18059 (2024): n. pag.
• Wang, Shuai, Ekaterina Khramtsova, Shengyao Zhuang and G. Zuccon. “FeB4RAG: Evaluating Federated Search in the Context of Retrieval Augmented Generation.” Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (2024): n. pag.
• Wang, Yu, Nedim Lipka, Ruiyi Zhang, Alexa F. Siu, Yuying Zhao, Bo Ni, Xin Wang, Ryan A. Rossi and Tyler Derr. “Augmenting Textual Generation via Topology Aware Retrieval.” ArXiv abs/2405.17602 (2024): n. pag.
• Wang, Zihao, Anji Liu, Haowei Lin, Jiaqi Li, Xiaojian Ma and Yitao Liang. “RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation.” ArXiv abs/2403.05313 (2024): n. pag.
• Khan, Ayman Asad, Md Toufique Hasan, Kai Kristian Kemell, Jussi Rasku, and Pekka Abrahamsson. "Developing retrieval augmented generation (RAG) based LLM systems from PDFs: an experience report." arXiv preprint arXiv:2410.15944 (2024).
• Bruch, Sebastian, Siyu Gai and Amir Ingber. “An Analysis of Fusion Functions for Hybrid Retrieval.” ACM Transactions on Information Systems 42 (2022): 1 - 35.
• Alavi, Maryam; Leidner, Dorothy E.; and Mousavi, Reza (2024) "A Knowledge Management Perspective of Generative Artificial Intelligence," Journal of the Association for Information Systems, 25(1), 1-12. DOI: 10.17705/1jais.00859. https://aisel.aisnet.org/jais/vol25/iss1/15 I. Blohm, F. Wortmann, C. Legner, and F. Köbler, “Data products, data mesh, and data fabric: New paradigm(s) for data and analytics?,” Bus. Inf. Syst. Eng., vol. 66, pp. 643–652, 2024 DOI: $10.1007/s12599-024-00876-5