Unstructured Data Analysis using LLMs: A Comprehensive Benchmark

TL;DR

This paper introduces UDA-Bench, a comprehensive benchmark with diverse datasets and queries, to evaluate large language model-based unstructured data analysis systems thoroughly.

Contribution

It provides the first large-scale, high-quality benchmark for evaluating UDA systems, including curated datasets, diverse queries, and detailed analysis of system components.

Findings

Identified key differences in query interfaces and optimization strategies.

Evaluated various UDA systems across multiple datasets and query types.

Provided insights into the strengths and weaknesses of existing approaches.

Abstract

Nowadays, the explosion of unstructured data presents immense analytical value. Leveraging the remarkable capability of large language models (LLMs) in extracting attributes of structured tables from unstructured data, researchers are developing LLM-powered data systems for users to analyze unstructured documents as working with a database. These unstructured data analysis (UDA) systems differ significantly in all aspects, including query interfaces, query optimization strategies, and operator implementations, making it unclear which performs best in which scenario. Unfortunately, there does not exist a comprehensive benchmark that offers high-quality, large-volume, and diverse datasets as well as rich query workload to thoroughly evaluate such systems. To fill this gap, we present UDA-Bench, the first benchmark for unstructured data analysis that meets all the above requirements. Specifically, we organize a team with 30 graduate students that spends over in total 10,000 hours on curating 5 datasets from various domains and constructing a relational database view from these datasets by manual annotation. These relational databases can be used as ground truth to evaluate any of these UDA systems despite their differences in programming interfaces. Moreover, we design diverse queries to analyze the attributes defined in the database schema, covering different types of analytical operators with varying selectivities and complexities. We conduct in-depth analysis of the key building blocks of existing UDA systems: query interface, query optimization, operator design, and data processing. We run exhaustive experiments over the benchmark to fully evaluate these systems and different techniques w.r.t. the above building blocks.