Open-domain question answering (QA) tasks usually require the retrieval of
relevant information from a large corpus to generate accurate answers. We
propose a novel approach called Generator-Retriever-Generator (GRG) that
combines document retrieval techniques with a large language model (LLM), by
first prompting the model to generate contextual documents based on a given
question. In parallel, a dual-encoder network retrieves documents that are
relevant to the question from an external corpus. The generated and retrieved
documents are then passed to the second LLM, which generates the final answer.
By combining document retrieval and LLM generation, our approach addresses the
challenges of open-domain QA, such as generating informative and contextually
relevant answers. GRG outperforms the state-of-the-art generate-then-read and
retrieve-then-read pipelines (GENREAD and RFiD) improving their performance at
least by +5.2, +4.2, and +1.6 on TriviaQA, NQ, and WebQ datasets, respectively.
We provide code, datasets, and checkpoints
\footnote{https://github.com/abdoelsayed2016/GRG}

通过结合文档检索技术与大型语言模型 (LLM)，我们提出了一种名为 Generator-Retriever-Generator (GRG) 的新方法，通过首先命令模型根据给定的问题生成上下文文档，然后使用双编码器网络从外部语料库中检索与问题相关的文档。将生成的和检索的文档传递给第二个 LLM 模型，生成最终的答案。通过结合文档检索和 LLM 生成，我们的方法解决了开放域问答的挑战，如生成信息丰富且上下文相关的答案。GRG 在 TriviaQA、NQ 和 WebQ 数据集上表现优于最先进的生成 - 阅读和检索 - 阅读流水线 (GENREAD 和 RFiD)，性能至少提高了 + 5.2、+4.2 和 + 1.6。我们提供代码、数据集和检查点。