PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision Makers

TL;DR

This paper introduces PlanRAG, a novel decision-making framework using large language models that generate plans and retrieve data iteratively, and presents a new benchmark for evaluating decision QA in complex scenarios.

Contribution

It proposes PlanRAG, a new iterative plan-then-retrieval method for decision making with LLMs, and introduces the Decision QA benchmark based on video game scenarios.

Findings

PlanRAG outperforms existing methods by 15.8% and 7.4% in two scenarios.

The Decision QA benchmark enables evaluation of decision-making capabilities.

The approach effectively combines planning and data retrieval for complex decision tasks.

Abstract

In this paper, we conduct a study to utilize LLMs as a solution for decision making that requires complex data analysis. We define Decision QA as the task of answering the best decision, $d_{best}$, for a decision-making question $Q$, business rules $R$ and a database $D$. Since there is no benchmark that can examine Decision QA, we propose Decision QA benchmark, DQA. It has two scenarios, Locating and Building, constructed from two video games (Europa Universalis IV and Victoria 3) that have almost the same goal as Decision QA. To address Decision QA effectively, we also propose a new RAG technique called the iterative plan-then-retrieval augmented generation (PlanRAG). Our PlanRAG-based LM generates the plan for decision making as the first step, and the retriever generates the queries for data analysis as the second step. The proposed method outperforms the state-of-the-art iterative RAG method by 15.8% in the Locating scenario and by 7.4% in the Building scenario, respectively. We release our code and benchmark at https://github.com/myeon9h/PlanRAG.