Domain-Specific Data Generation Framework for RAG Adaptation
This addresses the need for specialized training data in dynamic domains like scientific research and enterprise knowledge bases, but it is incremental as it builds on existing RAG adaptation approaches.
The paper tackles the problem of adapting Retrieval-Augmented Generation (RAG) systems to domain-specific settings by proposing RAGen, a scalable framework that generates domain-grounded question-answer-context triples, enabling efficient handling of large and evolving document corpora without redundant processing.
Retrieval-Augmented Generation (RAG) combines the language understanding and reasoning power of large language models (LLMs) with external retrieval to enable domain-grounded responses. Effectively adapting RAG systems to domain-specific settings requires specialized, context-rich training data beyond general-purpose question-answering. Here, we propose RAGen, a scalable and modular framework for generating domain-grounded question-answer-context (QAC) triples tailored to diverse RAG adaptation approaches. RAGen produces these QAC triples by identifying key concepts in documents, generating diverse questions guided by Bloom's Taxonomy-inspired principles, and pairing them with precise answers extracted from relevant contexts. RAGen supports multiple RAG adaptation strategies, including the optimization of key components such as the LLM, retriever, and embedding model, etc. Its modular pipeline features semantic chunking, hierarchical concept extraction, and multi-chunk retrieval, along with the introduction of curated distractor contexts to promote robust reasoning. Designed for scalability, RAGen efficiently handles large and evolving document corpora without redundant processing, making it especially suitable for dynamic evolving domains such as scientific research and enterprise knowledge bases.