解决检索中排除查询的反向问题，让系统真正忽略用户不要的内容。
EXCISE: Query-Side Exclusion for Late-Interaction Retrieval
- 在查询时识别排除主题并重嵌短列表，不修改索引
- 在六个数据集上将排除准确率从0.058提升至0.691
- 适合需要精准排除、不想微调模型的检索场景
晚交互检索器处理排除查询效果差。当用户请求X但不包含Z时，加性MaxSim得分会偏好包含Z的文档，这种现象称为排除反转。我们发现，仅靠对冻结向量的读出无法恢复约束，因为关键在于仅依赖查询本身识别被排除的主题。EXCISE在查询阶段运行，修正反转且不改变索引。两个共150万参数的查询侧模块识别主题并重嵌100篇文档的候选列表，一个无参数规则将匹配该主题的候选降权。在六个语料库和三种骨干模型上，EXCISE在全部十八个组合中均优于同骨干的冻结与微调基线。它将ExcluIR的排除成功@10从0.058提升至0.691，布尔非查询准确率从0.25–0.29提升至0.90–0.92。在1,860个查询上综合评估，其表现超越所有微调交叉编码器，而后者均损失无害nDCG@10，EXCISE则在最强骨干上保持与冻结基线一致。我们发布了X-BENCH，一个包含显式、隐式和复合排除的分层基准，包含无害与布尔控制条件。
原文摘要 · Abstract (English)
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.
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。