LAFA: Agentic LLM-Driven Federated Analytics over Decentralized Data Sources

TL;DR

LAFA is a novel system that combines large language models with federated analytics to enable natural language queries over decentralized data sources while preserving privacy, improving efficiency and success rates.

Contribution

LAFA introduces a hierarchical multi-agent architecture that translates natural language queries into optimized federated analytics workflows, supporting privacy-preserving data analysis.

Findings

LAFA outperforms baseline prompting strategies in success rates.

LAFA reduces redundant FA operations, saving computational resources.

LAFA effectively transforms natural language queries into executable FA workflows.

Abstract

Large Language Models (LLMs) have shown great promise in automating data analytics tasks by interpreting natural language queries and generating multi-operation execution plans. However, existing LLM-agent-based analytics frameworks operate under the assumption of centralized data access, offering little to no privacy protection. In contrast, federated analytics (FA) enables privacy-preserving computation across distributed data sources, but lacks support for natural language input and requires structured, machine-readable queries. In this work, we present LAFA, the first system that integrates LLM-agent-based data analytics with FA. LAFA introduces a hierarchical multi-agent architecture that accepts natural language queries and transforms them into optimized, executable FA workflows. A coarse-grained planner first decomposes complex queries into sub-queries, while a fine-grained planner maps each subquery into a Directed Acyclic Graph of FA operations using prior structural knowledge. To improve execution efficiency, an optimizer agent rewrites and merges multiple DAGs, eliminating redundant operations and minimizing computational and communicational overhead. Our experiments demonstrate that LAFA consistently outperforms baseline prompting strategies by achieving higher execution plan success rates and reducing resource-intensive FA operations by a substantial margin. This work establishes a practical foundation for privacy-preserving, LLM-driven analytics that supports natural language input in the FA setting.