On Evaluating the Integration of Reasoning and Action in LLM Agents with Database Question Answering

TL;DR

This paper presents a new dataset and evaluation framework for assessing how well Large Language Models can interact with SQL interpreters to answer complex database questions, revealing key challenges and proposing solutions.

Contribution

It introduces a novel long-form database question answering dataset, analyzes interaction bottlenecks, and develops a multi-agent peer-review evaluation framework for LLMs.

Findings

GPT-4 struggles with the task despite advanced capabilities.

Planning and multi-query generation are primary bottlenecks.

Multi-agent evaluation improves answer quality assessment.

Abstract

This study introduces a new long-form database question answering dataset designed to evaluate how Large Language Models (LLMs) interact with a SQL interpreter. The task necessitates LLMs to strategically generate multiple SQL queries to retrieve sufficient data from a database, to reason with the acquired context, and to synthesize them into a comprehensive analytical narrative. Our findings highlight that this task poses great challenges even for the state-of-the-art GPT-4 model. We propose and evaluate two interaction strategies, and provide a fine-grained analysis of the individual stages within the interaction. A key discovery is the identification of two primary bottlenecks hindering effective interaction: the capacity for planning and the ability to generate multiple SQL queries. To address the challenge of accurately assessing answer quality, we introduce a multi-agent evaluation framework that simulates the academic peer-review process, enhancing the precision and reliability of our evaluations. This framework allows for a more nuanced understanding of the strengths and limitations of current LLMs in complex retrieval and reasoning tasks.