Metraipolis › archive › dc_f8f779e5055e

Parameterized Dense-Sparse Fusion for Hybrid Retrieval: Tuning a Rank-Score Mix on BEIR SciFact with Qdrant — arXiv:2609.22770

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.22770
License (as archived)
CC0-1.0
Archived
2026-10-09T21:58:45.538Z
SHA-256 of the archived text
5f5a1ae03750f96caaba72a15a85e7de6e6d640a3f195f067a79de294e155bb7
Stable URL of this page
https://metraipolis.com/documento/dc_f8f779e5055e

How to cite

Parameterized Dense-Sparse Fusion for Hybrid Retrieval: Tuning a Rank-Score Mix on BEIR SciFact with Qdrant — arXiv:2609.22770 — arXiv (en), https://arxiv.org/abs/2609.22770 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f8f779e5055e (Metraipolis archive dc_f8f779e5055e, sha256 5f5a1ae03750f96caaba72a15a85e7de6e6d640a3f195f067a79de294e155bb7)
BibTeX
@misc{dcf8f779e5055e,
  title = {Parameterized Dense-Sparse Fusion for Hybrid Retrieval: Tuning a Rank-Score Mix on BEIR SciFact with Qdrant — arXiv:2609.22770},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.22770},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f8f779e5055e; sha256 5f5a1ae03750f96caaba72a15a85e7de6e6d640a3f195f067a79de294e155bb7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22770. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study a parameterized hybrid ranker that fuses a dense embedding list and a sparse lexical list. The method has a small, explicit parameter vector: a dense prior $\alpha \in [0,1]$, a score-versus-rank mix $\lambda \in [0,1]$, an RRF smoothing parameter $\kappa > 0$, optional list-geometry coefficients that move $\alpha$ per query, and a router margin $\tau$ that can turn sparse search off. We grid-search those ranges on SciFact train (809 queries) and freeze the chosen values on SciFact test (300). The tuned rank-score mix ($\alpha = 0.8$, $\lambda = 0.75$, $\kappa = 20$) reaches 0.753 nDCG@10 and 0.889 recall@10, outperforming dense BGE (0.742 / 0.871) and equal-weight RRF (0.707 nDCG@10) on that test split. A list-conditioned $\alpha$ adds +0.0006 nDCG; a sparse-off router is rejected by the same train split (any $\tau$ that skipped approximately 50% of queries lost nDCG). These coefficients are dataset-specific. Equal RRF with the same models does not beat dense on a nine-zip BEIR macro-average (0.479 vs. 0.519 nDCG@10). Repeating the same train-then-freeze sweep independently on all 20 indexed units beats equal RRF on 20/20 and dense on 16/20 (unit-mean nDCG@10 0.467 vs. 0.462 dense vs. 0.420 RRF). Other corpora should reuse the ranges, not a copy of the SciFact point.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.22770 — 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 2026-10-09T21:45:02.205Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).