MeMemo: On-device Retrieval Augmentation for Private and Personalized Text Generation

TL;DR

MeMemo is an open-source JavaScript toolkit that enables private, on-device retrieval-augmented text generation directly in browsers, addressing privacy concerns and expanding use cases for large language models.

Contribution

It introduces the first client-side dense retrieval toolkit using HNSW in JavaScript, facilitating private and personalized text generation without backend servers.

Findings

Enables efficient retrieval of millions of vectors in browsers.

Supports private, personalized content creation and prototyping.

Open-source and compatible with modern web technologies.

Abstract

Retrieval-augmented text generation (RAG) addresses the common limitations of large language models (LLMs), such as hallucination, by retrieving information from an updatable external knowledge base. However, existing approaches often require dedicated backend servers for data storage and retrieval, thereby limiting their applicability in use cases that require strict data privacy, such as personal finance, education, and medicine. To address the pressing need for client-side dense retrieval, we introduce MeMemo, the first open-source JavaScript toolkit that adapts the state-of-the-art approximate nearest neighbor search technique HNSW to browser environments. Developed with modern and native Web technologies, such as IndexedDB and Web Workers, our toolkit leverages client-side hardware capabilities to enable researchers and developers to efficiently search through millions of high-dimensional vectors in the browser. MeMemo enables exciting new design and research opportunities, such as private and personalized content creation and interactive prototyping, as demonstrated in our example application RAG Playground. Reflecting on our work, we discuss the opportunities and challenges for on-device dense retrieval. MeMemo is available at https://github.com/poloclub/mememo.