EXCISE：用于晚交互检索的查询端排除机制
EXCISE: Query-Side Exclusion for Late-Interaction Retrieval
浏览论文内容
中文总结 AI 辅助
EXCISE是一种查询端排除机制，通过两个查询端模块和无参数规则纠正晚交互检索的排除倒置问题，在多数据集、多骨干模型上性能优于基线，还发布了含多种排除类型的基准X-BENCH。
中文摘要 AI 辅助
晚交互检索器处理排除查询的效果很差。当用户询问X但排除Z时，加性MaxSim得分会倾向于覆盖Z的文档，我们将此问题称为“排除倒置”。我们表明，对冻结向量的任何读出操作都无法恢复该约束，因为难点在于识别被排除的主题，这仅取决于查询本身。EXCISE在查询时运行，在保持索引冻结的同时纠正了倒置问题。两个总计150万参数的查询端模块识别主题并重新嵌入100个文档的候选列表，一个无参数规则会降低匹配该主题的候选的排名。在六个数据集和三个骨干模型上，针对对应骨干模型自身的冻结和微调基线，EXCISE在所有18个骨干-数据集组合中都是性能最强的系统。它将ExcluIR上的排除成功@10从0.058提升至0.691，将布尔NOT准确率从0.25-0.29提升至0.90-0.92。在1860个查询的汇总结果中，它的性能优于每个微调交叉编码器，而每个微调交叉编码器都会损失无损害nDCG@10，EXCISE则在其最强骨干模型上与冻结基线表现相当。我们发布了X-BENCH，这是一个包含显式、隐式和复合排除的分层基准，带有无损害和布尔控制项。
英文摘要
Late-interaction retrievers handle exclusion queries poorly. When a user asks for X but not Z, the additive MaxSim score promotes documents covering Z, a problem we call exclusion inversion. We show that no readout of the frozen vectors recovers the constraint, because the difficulty lies in identifying the excluded topic, which depends on the query alone. EXCISE operates at query time and corrects the inversion while leaving the index frozen. Two query-side modules totalling 1.5M parameters identify the topic and re-embed a 100-document shortlist, and a parameter-free rule demotes candidates matching that topic. Across six collections and three backbones, EXCISE is the strongest system in all eighteen backbone-collection cells against that backbone's own frozen and fine-tuned baselines. It raises exclusion success@10 on ExcluIR from 0.058 to 0.691 and raises Boolean NOT accuracy from 0.25-0.29 to 0.90-0.92. Pooled over 1,860 queries, it outperforms every fine-tuned cross-encoder, each of which loses no-harm nDCG@10, whereas EXCISE matches its frozen baseline on its strongest backbone. We release X-BENCH, a tiered benchmark of explicit, implicit, and compound exclusions with no-harm and Boolean controls.
