对比LLM在跨语言检索中的表现，发现无需翻译也能有效检索。
Evaluating Large Language Models for Cross-Lingual Retrieval
- 用多语言双编码器做第一阶段检索，替代传统机器翻译。
- 强重排模型下翻译带来的提升明显减弱，甚至可省略。
- 指令微调的成对重排模型性能媲美列表重排，适合实际应用。
多阶段信息检索已成为搜索领域的主流范式。尽管大型语言模型（LLMs）已被广泛评估为单语检索中的第二阶段重排模型，但针对跨语言检索（CLIR）的大规模系统性比较仍缺乏。以往研究显示，基于LLM的重排模型能提升CLIR性能，但其评估依赖于以机器翻译（MT）作为第一阶段的词项检索，这不仅成本高昂，且易导致阶段间误差传播。我们在篇章级和文档级CLIR上的评估表明，采用多语言双编码器作为第一阶段检索器可进一步提升效果，且随着重排模型能力增强，翻译的益处逐渐减弱。我们还发现，基于指令微调的成对重排模型性能与列表重排模型相当。据我们所知，这是首个系统研究两阶段CLIR中检索器与重排器交互关系的工作。结果表明，若无机器翻译，当前最先进的重排模型在直接应用于CLIR时表现严重不足。
原文摘要 · Abstract (English)
Multi-stage information retrieval (IR) has become a widely-adopted paradigm in search. While Large Language Models (LLMs) have been extensively evaluated as second-stage reranking models for monolingual IR, a systematic large-scale comparison is still lacking for cross-lingual IR (CLIR). Moreover, while prior work shows that LLM-based rerankers improve CLIR performance, their evaluation setup relies on lexical retrieval with machine translation (MT) for the first stage. This is not only prohibitively expensive but also prone to error propagation across stages. Our evaluation on passage-level and document-level CLIR reveals that further gains can be achieved with multilingual bi-encoders as first-stage retrievers and that the benefits of translation diminishes with stronger reranking models. We further show that pairwise rerankers based on instruction-tuned LLMs perform competitively with listwise rerankers. To the best of our knowledge, we are the first to study the interaction between retrievers and rerankers in two-stage CLIR with LLMs. Our findings reveal that, without MT, current state-of-the-art rerankers fall severely short when directly applied in CLIR.
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。