Metraipolis › archive › dc_6093e50ed334

Query Expansion Should Be Coordinated: Dense Expands, Sparse Anchors — arXiv:2608.15851

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15851
License (as archived)
CC0-1.0
Archived
2026-10-18T19:01:17.488Z
SHA-256 of the archived text
5e810da5f774518e5b3dd4b07c8c475951009fdc4d7ff5945c33abc81655d5ca
Stable URL of this page
https://metraipolis.com/documento/dc_6093e50ed334

How to cite

Query Expansion Should Be Coordinated: Dense Expands, Sparse Anchors — arXiv:2608.15851 — arXiv (en), https://arxiv.org/abs/2608.15851 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6093e50ed334 (Metraipolis archive dc_6093e50ed334, sha256 5e810da5f774518e5b3dd4b07c8c475951009fdc4d7ff5945c33abc81655d5ca)
BibTeX
@misc{dc6093e50ed334,
  title = {Query Expansion Should Be Coordinated: Dense Expands, Sparse Anchors — arXiv:2608.15851},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15851},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6093e50ed334; sha256 5e810da5f774518e5b3dd4b07c8c475951009fdc4d7ff5945c33abc81655d5ca},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15851. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-augmented generation (RAG) systems rely on retrieval modules to ground large language model (LLM) outputs. LLM-based query expansion enriches retrieval with document-like passages, but evaluations of hybrid retrieval often fuse fixed top-L prefixes of dense and sparse rankings. Because L controls cross-channel contributions and ranking access, it can alter measured expansion gains. We therefore evaluate complete-list effectiveness and record per-channel replay stopping depths required to certify the ordered top-K. This changes the design: because both rankings determine the fused result, their query constructions should be coordinated rather than designed independently. We present DESA (Dense Expansion and Sparse Anchoring), which shares generated references across channels but specializes their integration. Orthogonal residual expansion adds new semantic directions to the dense query, whereas score-product anchoring reorders the original sparse support without admitting expansion-only matches. The same references thus play complementary roles: Dense expands; Sparse anchors. Across seven BEIR datasets, DESA improves nDCG@10 and Recall@20 over the unexpanded query by 3.82% and 2.38%, while reducing dense and sparse replay stopping depths by 36.90% and 36.56%.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.CL).
Categorie arXiv: Information Retrieval (cs.IR) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.15851 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T18:06:57.920Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).