Hierarchical Semantic Retrieval with Cobweb

arXiv:2510.02539v2 Announce Type: replace Abstract: 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.

Related

• Beyond Black-Box Interventions: Latent Probing for Faithful Retrieval-Augmented Generation
• Visual Late Chunking: An Empirical Study of Contextual Chunking for Efficient Visual Document Retrieval
• CoRoVA: Compressed Representations for Vector-Augmented Code Completion
• Chronological Knowledge Retrieval: A Retrieval-Augmented Generation Approach to Construction Project Documentation

Source: arXiv cs.CL | 2026-04-17