CAMPHOR: Collaborative Agents for Multi-input Planning and High-Order Reasoning On Device

TL;DR

CAMPHOR is a multi-agent on-device framework that enables small language models to perform complex reasoning and handle multiple inputs locally, improving privacy, latency, and efficiency in personalized mobile assistant tasks.

Contribution

We introduce CAMPHOR, a hierarchical multi-agent architecture with parameter sharing and prompt compression for efficient on-device reasoning and personalization.

Findings

Surpasses closed-source LLMs in task completion F1 by ~35%.

Eliminates server-device communication, enhancing privacy.

Reduces model size, latency, and memory usage significantly.

Abstract

While server-side Large Language Models (LLMs) demonstrate proficiency in function calling and complex reasoning, deploying Small Language Models (SLMs) directly on devices brings opportunities to improve latency and privacy but also introduces unique challenges for accuracy and memory. We introduce CAMPHOR, an innovative on-device SLM multi-agent framework designed to handle multiple user inputs and reason over personal context locally, ensuring privacy is maintained. CAMPHOR employs a hierarchical architecture where a high-order reasoning agent decomposes complex tasks and coordinates expert agents responsible for personal context retrieval, tool interaction, and dynamic plan generation. By implementing parameter sharing across agents and leveraging prompt compression, we significantly reduce model size, latency, and memory usage. To validate our approach, we present a novel dataset capturing multi-agent task trajectories centered on personalized mobile assistant use-cases. Our experiments reveal that fine-tuned SLM agents not only surpass closed-source LLMs in task completion F1 by~35\% but also eliminate the need for server-device communication, all while enhancing privacy.