Satyrn: A Platform for Analytics Augmented Generation

TL;DR

Satyrn is a neurosymbolic platform that enhances large language model report generation by integrating structured data analysis, achieving high accuracy and coherence even with smaller models.

Contribution

The paper introduces Analytics Augmented Generation (AAG) and the Satyrn platform, combining structured data analysis with LLMs for improved factual accuracy in reports.

Findings

Over 86% claim accuracy in generated reports

Satyrn outperforms GPT-4 Code Interpreter in claim accuracy

Effective with smaller language models like Mistral-7B

Abstract

Large language models (LLMs) are capable of producing documents, and retrieval augmented generation (RAG) has shown itself to be a powerful method for improving accuracy without sacrificing fluency. However, not all information can be retrieved from text. We propose an approach that uses the analysis of structured data to generate fact sets that are used to guide generation in much the same way that retrieved documents are used in RAG. This analytics augmented generation (AAG) approach supports the ability to utilize standard analytic techniques to generate facts that are then converted to text and passed to an LLM. We present a neurosymbolic platform, Satyrn, that leverages AAG to produce accurate, fluent, and coherent reports grounded in large scale databases. In our experiments, we find that Satyrn generates reports in which over 86% of claims are accurate while maintaining high levels of fluency and coherence, even when using smaller language models such as Mistral-7B, as compared to GPT-4 Code Interpreter in which just 57% of claims are accurate.