zELO：基于ELO启发的重排序器与嵌入模型训练方法

摘要

我们提出了一种名为zELO的创新训练方法，该方法通过分析排序任务与Thurstone模型的静态等价性来优化检索性能。基于zELO方法，我们利用无监督数据训练了一套最先进的开放权重重排序模型：zerank-1和zerank-1-small。这些模型在多个领域（包括金融、法律、代码和STEM）中均取得了最高的检索分数，在NDCG@10和召回率上均超越了闭源专有重排序器。这些模型还展现了极强的泛化能力，在跨领域和私有客户数据集上保持了零样本性能。训练数据包含112,000个查询，每个查询对应100篇文档，从未标注的查询和文档端到端训练完成，耗时不到10,000个H100小时。

English

We introduce a novel training methodology named zELO, which optimizes retrieval performance via the analysis that ranking tasks are statically equivalent to a Thurstone model. Based on the zELO method, we use unsupervised data in order train a suite of state-of-the-art open-weight reranker models: zerank-1 and zerank-1-small. These models achieve the highest retrieval scores in multiple domains, including finance, legal, code, and STEM, outperforming closed-source proprietary rerankers on both NDCG@10 and Recall. These models also demonstrate great versatility, maintaining their 0-shot performance on out-of-domain and private customer datasets. The training data included 112,000 queries and 100 documents per query, and was trained end-to-end from unannotated queries and documents in less than 10,000 H100-hours.

zELO：基于ELO启发的重排序器与嵌入模型训练方法

zELO: ELO-inspired Training Method for Rerankers and Embedding Models

摘要

Support