Zero-RAG: Towards Retrieval-Augmented Generation with Zero Redundant Knowledge
This addresses efficiency and performance issues in RAG systems for users deploying LLMs with external knowledge bases, though it appears incremental.
The paper tackles the problem of knowledge redundancy between LLMs' internal knowledge and external corpora in Retrieval-Augmented Generation, which increases retrieval costs and hurts performance. The proposed Zero-RAG method prunes 30% of the Wikipedia corpus, accelerates retrieval by 22%, and maintains RAG performance.
Retrieval-Augmented Generation has shown remarkable results to address Large Language Models' hallucinations, which usually uses a large external corpus to supplement knowledge to LLMs. However, with the development of LLMs, the internal knowledge of LLMs has expanded significantly, thus causing significant knowledge redundancy between the external corpus and LLMs. On the one hand, the indexing cost of dense retrieval is highly related to the corpus size and thus significant redundant knowledge intensifies the dense retrieval's workload. On the other hand, the redundant knowledge in the external corpus is not helpful to LLMs and our exploratory analysis shows that it instead hurts the RAG performance on those questions which the LLM can answer by itself. To address these issues, we propose Zero-RAG to tackle these challenges. Specifically, we first propose the Mastery-Score metric to identify redundant knowledge in the RAG corpus to prune it. After pruning, answers to "mastered" questions rely primarily on internal knowledge of the LLM. To better harness the internal capacity, we propose Query Router and Noise-Tolerant Tuning to avoid the irrelevant documents' distraction and thus further improve the LLM's utilization of internal knowledge with pruned corpus. Experimental results show that Zero-RAG prunes the Wikipedia corpus by 30\% and accelerates the retrieval stage by 22\%, without compromising RAG's performance.