Benchmarking Retrieval-Augmented Generation for Chemistry
By: Xianrui Zhong , Bowen Jin , Siru Ouyang and more
Potential Business Impact:
Helps computers answer chemistry questions better.
Retrieval-augmented generation (RAG) has emerged as a powerful framework for enhancing large language models (LLMs) with external knowledge, particularly in scientific domains that demand specialized and dynamic information. Despite its promise, the application of RAG in the chemistry domain remains underexplored, primarily due to the lack of high-quality, domain-specific corpora and well-curated evaluation benchmarks. In this work, we introduce ChemRAG-Bench, a comprehensive benchmark designed to systematically assess the effectiveness of RAG across a diverse set of chemistry-related tasks. The accompanying chemistry corpus integrates heterogeneous knowledge sources, including scientific literature, the PubChem database, PubMed abstracts, textbooks, and Wikipedia entries. In addition, we present ChemRAG-Toolkit, a modular and extensible RAG toolkit that supports five retrieval algorithms and eight LLMs. Using ChemRAG-Toolkit, we demonstrate that RAG yields a substantial performance gain -- achieving an average relative improvement of 17.4% over direct inference methods. We further conduct in-depth analyses on retriever architectures, corpus selection, and the number of retrieved passages, culminating in practical recommendations to guide future research and deployment of RAG systems in the chemistry domain. The code and data is available at https://chemrag.github.io.
Similar Papers
mmRAG: A Modular Benchmark for Retrieval-Augmented Generation over Text, Tables, and Knowledge Graphs
Information Retrieval
Tests AI that uses pictures and words.
A Survey on Knowledge-Oriented Retrieval-Augmented Generation
Computation and Language
Lets computers use outside facts to answer questions.
A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models
Computation and Language
Helps computers understand complex topics better.