Search papers, labs, and topics across Lattice.
The paper introduces the Graph Data Science (GDS) agent, a system that augments LLMs with graph algorithms as tools for reasoning over graph-structured data. This addresses the limitation of LLMs in processing and reasoning about large-scale graphs by providing a comprehensive set of graph algorithms accessible via function calling. Experiments using new benchmarks demonstrate the GDS agent's ability to solve a variety of graph tasks and provide accurate, grounded answers to questions requiring graph algorithmic reasoning.
LLMs can now reason about complex graph data with high accuracy, thanks to a new agent that integrates graph algorithms as callable tools.
Large language models (LLMs) have shown remarkable multimodal information processing and reasoning ability. When equipped with tools through function calling and enhanced with retrieval-augmented techniques, compound LLM-based systems can access closed data sources and answer questions about them. However, they still struggle to process and reason over large-scale graph-structure data. We introduce the GDS (Graph Data Science) agent in this technical report. The GDS agent introduces a comprehensive set of graph algorithms as tools, together with preprocessing (retrieval) and postprocessing of algorithm results, in a model context protocol (MCP) server. The server can be used with any modern LLM out-of-the-box. GDS agent allows users to ask any question that implicitly and intrinsically requires graph algorithmic reasoning about their data, and quickly obtain accurate and grounded answers. We introduce new benchmarks that evaluate intermediate tool calls as well as final responses. The results indicate that GDS agent is able to solve a wide spectrum of graph tasks. We also provide detailed case studies for more open-ended tasks and study scenarios where the agent struggles. Finally, we discuss the remaining challenges and the future roadmap.