Control Token with Dense Passage Retrieval
This addresses retrieval accuracy issues in RAG for domain-specific queries, but it is incremental as it builds on existing DPR and RAG methods.
The study tackled hallucination in large language models by enhancing Dense Passage Retrieval with control tokens, achieving a 13% improvement in Top-1 accuracy and 4% in Top-20 accuracy over the standard DPR model.
This study addresses the hallucination problem in large language models (LLMs). We adopted Retrieval-Augmented Generation(RAG) (Lewis et al., 2020), a technique that involves embedding relevant information in the prompt to obtain accurate answers. However, RAG also faced inherent issues in retrieving correct information. To address this, we employed the Dense Passage Retrieval(DPR) (Karpukhin et al., 2020) model for fetching domain-specific documents related to user queries. Despite this, the DPR model still lacked accuracy in document retrieval. We enhanced the DPR model by incorporating control tokens, achieving significantly superior performance over the standard DPR model, with a 13% improvement in Top-1 accuracy and a 4% improvement in Top-20 accuracy.