Metraipolis › archive › dc_ea75e0b38405

EXCISE: Query-Side Exclusion for Late-Interaction Retrieval — arXiv:2608.05497

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.05497
License (as archived)
CC0-1.0
Archived
2026-10-31T06:30:06.459Z
SHA-256 of the archived text
cd1d9bc9b3fec2cbab12b009d22182d6995ff4559ddb3cd71292eab3e307a6e7
Stable URL of this page
https://metraipolis.com/documento/dc_ea75e0b38405

How to cite

EXCISE: Query-Side Exclusion for Late-Interaction Retrieval — arXiv:2608.05497 — arXiv (en), https://arxiv.org/abs/2608.05497 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea75e0b38405 (Metraipolis archive dc_ea75e0b38405, sha256 cd1d9bc9b3fec2cbab12b009d22182d6995ff4559ddb3cd71292eab3e307a6e7)
BibTeX
@misc{dcea75e0b38405,
  title = {EXCISE: Query-Side Exclusion for Late-Interaction Retrieval — arXiv:2608.05497},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05497},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea75e0b38405; sha256 cd1d9bc9b3fec2cbab12b009d22182d6995ff4559ddb3cd71292eab3e307a6e7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05497. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Late-interaction retrievers handle exclusion queries poorly. When a user asks for X but not Z, the additive MaxSim score promotes documents covering Z, a problem we call exclusion inversion. We show that no readout of the frozen vectors recovers the constraint, because the difficulty lies in identifying the excluded topic, which depends on the query alone. EXCISE operates at query time and corrects the inversion while leaving the index frozen. Two query-side modules totalling 1.5M parameters identify the topic and re-embed a 100-document shortlist, and a parameter-free rule demotes candidates matching that topic. Across six collections and three backbones, EXCISE is the strongest system in all eighteen backbone-collection cells against that backbone's own frozen and fine-tuned baselines. It raises exclusion success@10 on ExcluIR from 0.058 to 0.691 and raises Boolean NOT accuracy from 0.25-0.29 to 0.90-0.92. Pooled over 1,860 queries, it outperforms every fine-tuned cross-encoder, each of which loses no-harm nDCG@10, whereas EXCISE matches its frozen baseline on its strongest backbone. We release X-BENCH, a tiered benchmark of explicit, implicit, and compound exclusions with no-harm and Boolean controls.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2608.05497 — 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 2026-10-31T06:15:18.494Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).