Dense retrieval models have predominantly been studied for English, where
models have shown great success, due to the availability of human-labeled
training pairs. However, there has been limited success for multilingual
retrieval so far, as training data is uneven or scarcely available across
multiple languages. Synthetic training data generation is promising (e.g.,
InPars or Promptagator), but has been investigated only for English. Therefore,
to study model capabilities across both cross-lingual and monolingual retrieval
tasks, we develop SWIM-IR, a synthetic retrieval training dataset containing 33
(high to very-low resource) languages for training multilingual dense retrieval
models without requiring any human supervision. To construct SWIM-IR, we
propose SAP (summarize-then-ask prompting), where the large language model
(LLM) generates a textual summary prior to the query generation step. SAP
assists the LLM in generating informative queries in the target language. Using
SWIM-IR, we explore synthetic fine-tuning of multilingual dense retrieval
models and evaluate them robustly on three retrieval benchmarks: XOR-Retrieve
(cross-lingual), XTREME-UP (cross-lingual) and MIRACL (monolingual). Our
models, called SWIM-X, are competitive with human-supervised dense retrieval
models, e.g., mContriever, finding that SWIM-IR can cheaply substitute for
expensive human-labeled retrieval training data.

通过使用 SWIM-IR 合成训练数据集，我们研究了多语言密集检索模型的能力，并在三个检索基准上对其进行了全面评估，发现 SWIM-IR 可以以较低成本替代昂贵的人工标记检索训练数据。