发表机构
Los Alamos National Laboratory（洛斯阿拉莫斯国家实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过受控实验证明，生成式查询扩展在强稀疏检索器SPLADE-v3上仍能显著提升检索性能，最佳相对增益达9.47%，且原始查询权重不低于30%时增益稳定，表明生成词汇可有效补充强检索器。
AI 中文摘要
科学查询通常简短，而相关论文使用专业词汇。生成式查询扩展可以弥合这一不匹配，但早期研究表明，随着底层检索器变得更强，其价值会缩小。我们测试了四种生成格式：术语列表、伪文档、多个伪参考文献和语料库引导文本，全部与SPLADE-v3在NFCorpus、TREC-COVID和SciDocs上结合使用。每个条件都搜索相同的冻结文档索引，并遵循相同的查询侧集成规则和256维预算，从而隔离了添加内容的影响。所有十二种方法-集合比较都改善了聚合nDCG@10，最佳相对增益分别为4.81%、8.92%和9.47%。在Holm校正后，有十一种仍然显著。在114种插值设置中，增益在103种中持续存在，包括所有将至少30%的混合权重分配给原始查询的设置。打乱文本和非上下文词汇袋对照在所有24种聚合比较中也保持在基线之上，表明添加的词汇带来了大部分收益。相比之下，语料库诱导的类型化概念图没有产生一致的增益，其关系、深度、验证、随机和门控控制也无法挽救它。因此，生成的词汇可以补充强学习型稀疏检索器，前提是原始查询仍然被强烈表示。
英文摘要
Scientific queries are often brief, while relevant papers use specialized vocabulary. Generated query expansion can bridge this mismatch, but earlier work suggests that its value shrinks as the underlying retriever becomes stronger. We test the four generated formats of term lists, a pseudo-document, multiple pseudo-references, and corpus-steered text all together with SPLADE-v3 on NFCorpus, TREC-COVID, and SciDocs. Every condition searches the same frozen document index and follows the same query-side integration rule and 256-dimension budget, isolating the effect of the added content. All twelve method-collection comparisons improve aggregate nDCG@10, with best relative gains of 4.81%, 8.92%, and 9.47%. Eleven remain significant after Holm correction. The gain persists in 103 of 114 interpolation settings, including every setting that assigns at least 30% of the mixture weight to the original query. Shuffled-text and non-contextual lexical-bag controls also remain above baseline in all 24 aggregate comparisons, showing that the added vocabulary carries most of the benefit. A corpus-induced typed concept graph, by contrast, produces no consistent gain, and its relation, depth, validation, random, and gating controls do not rescue it. Generated vocabulary can therefore complement a strong learned sparse retriever, provided that the original query remains strongly represented.
Comments8 pages, 5 tables, 3 figures
