Neural document retrieval often treats a corpus as a flat cloud of vectors scored at a single granularity, leaving corpus structure underused and explanations opaque. We use Cobweb--a hierarchy-aware framework--to organize sentence embeddings into a prototype tree and rank documents via coarse-to-fine traversal. Internal nodes act as concept prototypes, providing multi-granular relevance signals and a transparent rationale through retrieval paths. We instantiate two inference approaches: a generalized best-first search and a lightweight path-sum ranker. We evaluate our approaches on MS MARCO and QQP with encoder (e.g., BERT/T5) and decoder (GPT-2) representations. Our results show that our retrieval approaches match the dot product search on strong encoder embeddings while remaining robust when kNN degrades: with GPT-2 vectors, dot product performance collapses whereas our approaches still retrieve relevant results. Overall, our experiments suggest that Cobweb provides competitive effectiveness, improved robustness to embedding quality, scalability, and interpretable retrieval via hierarchical prototypes.

翻译：神经文档检索通常将语料库视为一个扁平的向量云，并在单一粒度下进行评分，这导致语料库结构未被充分利用，且检索结果的解释性不足。我们采用Cobweb——一种层级感知框架——将句子嵌入组织成原型树，并通过由粗到细的遍历对文档进行排序。内部节点充当概念原型，提供多粒度相关性信号，并通过检索路径提供透明的推理依据。我们实例化了两种推理方法：一种广义的最佳优先搜索和一种轻量化的路径总和排名器。我们在MS MARCO和QQP数据集上，使用编码器（例如BERT/T5）和解码器（GPT-2）的表示方法评估了我们的方法。结果表明，我们的检索方法在强编码器嵌入上可与点积搜索匹敌，同时在kNN性能下降时仍保持稳健：使用GPT-2向量时，点积检索性能崩溃，而我们的方法仍能检索到相关结果。总体而言，我们的实验表明，Cobweb通过层级原型提供了具有竞争力的有效性、对嵌入质量的改进鲁棒性、可扩展性以及可解释的检索。