arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2504.12972cs.CL
收藏

混合检索增强型多文档摘要的最优上下文长度估计

Estimating Optimal Context Length for Hybrid Retrieval-augmented Multi-document Summarization

Carnegie Mellon University(卡内基梅隆大学)

Adithya Pratapa, Teruko Mitamura

AI总结:

针对检索增强多文档摘要中检索上下文长度敏感问题,提出结合检索增强系统与长上下文窗口的混合方法,通过银标参考估计最优长度,在多模型上有效且泛化性强。

摘要:

语言模型长上下文推理能力的最新进展推动了大规模多文档摘要领域的有趣应用。然而,先前研究表明这些长上下文模型在其宣称的上下文窗口内效果不佳。为此,检索增强系统提供了高效且有效的替代方案,但其性能对检索上下文长度的选择高度敏感。本研究提出一种混合方法,将检索增强系统与近期语言模型支持的长上下文窗口相结合。该方法首先将最优检索长度估计为检索器、摘要器和数据集的函数;在数据集的随机抽样子集上,使用大语言模型(LLM)面板生成银标参考池;利用这些银标参考估计给定检索增强生成(RAG)系统配置的最优上下文长度。多文档摘要任务的结果表明,该方法在不同模型类别和规模上均有效;与RULER、HELMET等强长上下文基准的长度估计相比表现优异;分析还突显了该估计方法对超长上下文语言模型的有效性及其对新型语言模型的泛化能力。

英文摘要:

Recent advances in long-context reasoning abilities of language models led to interesting applications in large-scale multi-document summarization. However, prior work has shown that these long-context models are not effective at their claimed context windows. To this end, retrieval-augmented systems provide an efficient and effective alternative. However, their performance can be highly sensitive to the choice of retrieval context length. In this work, we present a hybrid method that combines retrieval-augmented systems with long-context windows supported by recent language models. Our method first estimates the optimal retrieval length as a function of the retriever, summarizer, and dataset. On a randomly sampled subset of the dataset, we use a panel of LLMs to generate a pool of silver references. We use these silver references to estimate the optimal context length for a given RAG system configuration. Our results on the multi-document summarization task showcase the effectiveness of our method across model classes and sizes. We compare against length estimates from strong long-context benchmarks such as RULER and HELMET. Our analysis also highlights the effectiveness of our estimation method for very long-context LMs and its generalization to new classes of LMs.

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。

↑