arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2201.10582cs.IR
收藏

域外语义来拯救!Zero-Shot混合检索模型

Out-of-Domain Semantics to the Rescue! Zero-Shot Hybrid Retrieval Models

Tao Chen, Mingyang Zhang, Jing Lu, Michael Bendersky, Marc Najork

摘要:

基于预训练语言模型(例如,BERT)的深度检索模型在许多段落检索任务中取得了优于词汇检索模型(例如,BM 25)的性能。然而,有限的工作已经做了推广到其他任务和领域的深度检索模型。在这项工作中,我们仔细选择了五个数据集,包括两个域内数据集和三个具有不同程度的域偏移的域外数据集,并研究了深度模型在zero-shot设置下的泛化。我们的研究结果表明,当目标域与模型训练的源域非常不同时,深度检索模型的性能会显著恶化。相反,词汇模型在跨域中更鲁棒。因此,我们提出了一个简单而有效的框架来整合词汇和深度检索模型。我们的实验表明,这两个模型是互补的,即使当深层模型在域外设置较弱。在三个域外数据集上,混合模型比深度检索模型平均获得20.4%的相对增益,比词汇模型平均获得9.54%的相对增益。

英文摘要:

The pre-trained language model (eg, BERT) based deep retrieval models achieved superior performance over lexical retrieval models (eg, BM25) in many passage retrieval tasks. However, limited work has been done to generalize a deep retrieval model to other tasks and domains. In this work, we carefully select five datasets, including two in-domain datasets and three out-of-domain datasets with different levels of domain shift, and study the generalization of a deep model in a zero-shot setting. Our findings show that the performance of a deep retrieval model is significantly deteriorated when the target domain is very different from the source domain that the model was trained on. On the contrary, lexical models are more robust across domains. We thus propose a simple yet effective framework to integrate lexical and deep retrieval models. Our experiments demonstrate that these two models are complementary, even when the deep model is weaker in the out-of-domain setting. The hybrid model obtains an average of 20.4% relative gain over the deep retrieval model, and an average of 9.54% over the lexical model in three out-of-domain datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑