Evaluating Self-Generated Documents for Enhancing Retrieval-Augmented Generation with Large Language Models
This work addresses the challenge of optimizing RAG systems for better knowledge-intensive QA, though it appears incremental as it builds on existing Self-Docs research by exploring their properties and combination strategies.
The paper tackles the problem of enhancing retrieval-augmented generation (RAG) systems by evaluating self-generated documents (Self-Docs) from large language models, finding that certain types of Self-Docs significantly improve performance in knowledge-intensive question answering tasks.
The integration of documents generated by LLMs themselves (Self-Docs) alongside retrieved documents has emerged as a promising strategy for retrieval-augmented generation systems. However, previous research primarily focuses on optimizing the use of Self-Docs, with their inherent properties remaining underexplored. To bridge this gap, we first investigate the overall effectiveness of Self-Docs, identifying key factors that shape their contribution to RAG performance (RQ1). Building on these insights, we develop a taxonomy grounded in Systemic Functional Linguistics to compare the influence of various Self-Docs categories (RQ2) and explore strategies for combining them with external sources (RQ3). Our findings reveal which types of Self-Docs are most beneficial and offer practical guidelines for leveraging them to achieve significant improvements in knowledge-intensive question answering tasks.