arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.01263cs.IR
收藏

使用前向索引和轻量级编码器的高效神经排序

Efficient Neural Ranking using Forward Indexes and Lightweight Encoders

Jurek Leonhardt, Henrik Müller, Koustav Rudra, Megha Khosla, Abhijit Anand, Avishek Anand

AI总结:

本文提出 Fast-Forward 向量前向索引,结合词法与语义分数插值、表示预计算、轻量编码器及动态标记丢弃,实现高深度、低延迟且无需 GPU 的高效神经重排序。

摘要:

基于双编码器的密集检索模型已成为信息检索(IR)领域的标准。它们采用基于 Transformer 的大型语言模型,而这类模型在资源消耗和延迟方面众所周知地低效。我们提出 Fast-Forward 索引——一种向量前向索引,利用双编码器模型的语义匹配能力实现高效且有效的重排序。我们的框架支持在极高检索深度下进行重排序,并通过分数插值结合词法匹配与语义匹配的优点。此外,为缓解双编码器的局限,我们应对两大挑战:第一,通过预计算表示以完全避免不必要的计算,或降低编码器复杂度,来提升计算效率。这使我们能够显著改善排序效率和延迟。第二,我们优化索引的内存占用和维护成本;提出两种互补技术来减小索引大小,并表明通过动态丢弃无关文档标记,可大幅提升索引维护效率。我们进行评估以证明 Fast-Forward 索引的有效性和效率——我们的方法具有低延迟,并且无需 GPU 等硬件加速即可取得具有竞争力的结果。

英文摘要:

Dual-encoder-based dense retrieval models have become the standard in IR. They employ large Transformer-based language models, which are notoriously inefficient in terms of resources and latency. We propose Fast-Forward indexes -- vector forward indexes which exploit the semantic matching capabilities of dual-encoder models for efficient and effective re-ranking. Our framework enables re-ranking at very high retrieval depths and combines the merits of both lexical and semantic matching via score interpolation. Furthermore, in order to mitigate the limitations of dual-encoders, we tackle two main challenges: Firstly, we improve computational efficiency by either pre-computing representations, avoiding unnecessary computations altogether, or reducing the complexity of encoders. This allows us to considerably improve ranking efficiency and latency. Secondly, we optimize the memory footprint and maintenance cost of indexes; we propose two complementary techniques to reduce the index size and show that, by dynamically dropping irrelevant document tokens, the index maintenance efficiency can be improved substantially. We perform evaluation to show the effectiveness and efficiency of Fast-Forward indexes -- our method has low latency and achieves competitive results without the need for hardware acceleration, such as GPUs.

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。

↑