提出按语义分组的解释方法，让文档排序原因更贴近真实语义。
Listwise Explanation of Embedding-Based Rankings via Semantic Chunk Grouping
- 将相关语义片段聚类为共享特征，提升解释粒度
- 在多个数据集上验证，不同模型适合不同解释单元
- 适合关注排序逻辑与语义一致性研究者
密集嵌入排序器通过上下文相关的句子级和段落级表示对文档打分。然而，许多列表式解释方法仍依赖孤立词汇进行归因，这种特征粒度与密集语义表示不匹配，导致解释碎片化。本文提出ChunkGroupSHAP，一种列表式Shapley方法，将语义相关的文本块聚类为跨文档共享特征。掩码一个分组会同时影响所有包含相关证据的文档，使归因粒度更接近密集表示，同时保持列表式设定。在MS MARCO、FinanceBench、AILACaseDocs和FinQA四个数据集上，使用E5排序器和BM25的实验表明，最佳解释单元取决于场景：词级特征适用于基于词法的BM25，语料级分组适用于密集排序器，查询局部分组则适用于异构网络检索。因此，解释单元应匹配排序器的表征粒度和检索语料结构。
原文摘要 · Abstract (English)
Dense embedding rankers score documents through contextual sentence- and passage-level representations. Yet many listwise explanation methods still attribute rankings to isolated words. This feature-unit mismatch leaves word-level features too fragmented for dense semantic ranking. We introduce ChunkGroupSHAP, a listwise Shapley method that clusters semantically related chunks into shared cross-document features. Masking a group perturbs all documents with related evidence, attributing rankings at a granularity closer to dense representations while preserving the listwise setup. Our findings across MS MARCO, FinanceBench, AILACaseDocs, and FinQA with E5 rankers and BM25 show that the best explanation unit is setting-dependent: word features for lexical BM25, corpus-level groups for dense rankers, and query-local grouping for heterogeneous web retrieval. Feature units should thus follow both the ranker's representational granularity and the structure of the retrieved corpus.
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。