BiomedRAG: A Retrieval Augmented Large Language Model for Biomedicine
This work addresses the need for more reliable and accurate AI tools in biomedicine and healthcare, though it is incremental as it builds on existing retrieval-augmented methods with a simpler design.
The paper tackles the problem of inaccuracies and hallucinations in large language models for biomedical applications by introducing BiomedRAG, a retrieval-augmented approach that directly inputs retrieved documents into the LLM, achieving superior performance across 5 biomedical NLP tasks, such as triple extraction with micro-F1 scores of 81.42 and 88.83 on specific datasets.
Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains; however, these models encounter issues such as generating inaccurate information or hallucinations. Retrieval-augmented generation provided a solution for these models to update knowledge and enhance their performance. In contrast to previous retrieval-augmented LMs, which utilize specialized cross-attention mechanisms to help LLM encode retrieved text, BiomedRAG adopts a simpler approach by directly inputting the retrieved chunk-based documents into the LLM. This straightforward design is easily applicable to existing retrieval and language models, effectively bypassing noise information in retrieved documents, particularly in noise-intensive tasks. Moreover, we demonstrate the potential for utilizing the LLM to supervise the retrieval model in the biomedical domain, enabling it to retrieve the document that assists the LM in improving its predictions. Our experiments reveal that with the tuned scorer,\textsc{ BiomedRAG} attains superior performance across 5 biomedical NLP tasks, encompassing information extraction (triple extraction, relation extraction), text classification, link prediction, and question-answering, leveraging over 9 datasets. For instance, in the triple extraction task, \textsc{BiomedRAG} outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively.