H+ Embedding: Harmonizing Global and Token-Level Retrieval with Context-Dependent Phrases

본문 미리보기

arXiv:2608.00065v1 Announce Type: new Abstract: Terminology-intensive retrieval, especially in medical settings, depends on preserving multi-word entities, abbreviations, numerical constraints, and compositional concepts. However, existing representations lie at two extremes: single-vector retrievers often over-compress local relevance signals, while token-level late interaction retains every tokenizer subword at substantial indexing, storage, and scoring cost. This mismatch raises a natural qu