HiPerRAG: High-Performance Retrieval Augmented Generation for Scientific Insights

TL;DR

HiPerRAG leverages high-performance computing to enable scalable retrieval-augmented generation from over 3.6 million scientific articles, significantly improving scientific question answering accuracy and facilitating cross-disciplinary research.

Contribution

The paper introduces HiPerRAG, a novel HPC-powered RAG workflow with new models Oreo and ColTrast, designed to handle large-scale scientific knowledge retrieval efficiently.

Findings

Achieves 90% accuracy on SciQ benchmark

Attains 76% accuracy on PubMedQA benchmark

Outperforms domain-specific models and GPT-4 in scientific QA

Abstract

The volume of scientific literature is growing exponentially, leading to underutilized discoveries, duplicated efforts, and limited cross-disciplinary collaboration. Retrieval Augmented Generation (RAG) offers a way to assist scientists by improving the factuality of Large Language Models (LLMs) in processing this influx of information. However, scaling RAG to handle millions of articles introduces significant challenges, including the high computational costs associated with parsing documents and embedding scientific knowledge, as well as the algorithmic complexity of aligning these representations with the nuanced semantics of scientific content. To address these issues, we introduce HiPerRAG, a RAG workflow powered by high performance computing (HPC) to index and retrieve knowledge from more than 3.6 million scientific articles. At its core are Oreo, a high-throughput model for multimodal document parsing, and ColTrast, a query-aware encoder fine-tuning algorithm that enhances retrieval accuracy by using contrastive learning and late-interaction techniques. HiPerRAG delivers robust performance on existing scientific question answering benchmarks and two new benchmarks introduced in this work, achieving 90% accuracy on SciQ and 76% on PubMedQA-outperforming both domain-specific models like PubMedGPT and commercial LLMs such as GPT-4. Scaling to thousands of GPUs on the Polaris, Sunspot, and Frontier supercomputers, HiPerRAG delivers million document-scale RAG workflows for unifying scientific knowledge and fostering interdisciplinary innovation.