用无监督重排序融合生成与检索知识，提升开放域问答效果
Improving Generated and Retrieved Knowledge Combination Through Zero-shot Generation
- 通过双重重排序分别优化检索和生成的文本
- 在NQ和WebQ上分别提升1.7和1.6分，TriviaQA表现相当
- 无需标签，适合缺乏标注数据的问答系统
开放域问答通过结合忠实检索的段落和大语言模型生成的相关段落，取得了显著进展。然而，这些知识来源缺乏明确的配对标签。为此，我们提出一种无监督且简洁的框架BRMGR（Bi-Reranking for Merging Generated and Retrieved Knowledge），利用重排序方法分别处理检索段落和生成段落。通过两种独立的重排序机制配对两类段落，并采用贪心匹配进行融合。我们证明，BRMGR等价于施加二分图匹配损失，为每个检索段落分配一个对应的生成段落。在三个数据集上的实验表明，该模型在NQ和WebQ上分别提升了1.7和1.6分，在TriviaQA上表现接近现有最优基线。
原文摘要 · Abstract (English)
Open-domain Question Answering (QA) has garnered substantial interest by combining the advantages of faithfully retrieved passages and relevant passages generated through Large Language Models (LLMs). However, there is a lack of definitive labels available to pair these sources of knowledge. In order to address this issue, we propose an unsupervised and simple framework called Bi-Reranking for Merging Generated and Retrieved Knowledge (BRMGR), which utilizes re-ranking methods for both retrieved passages and LLM-generated passages. We pair the two types of passages using two separate re-ranking methods and then combine them through greedy matching. We demonstrate that BRMGR is equivalent to employing a bipartite matching loss when assigning each retrieved passage with a corresponding LLM-generated passage. The application of our model yielded experimental results from three datasets, improving their performance by +1.7 and +1.6 on NQ and WebQ datasets, respectively, and obtaining comparable result on TriviaQA dataset when compared to competitive baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。