Metraipolis › archive › dc_ff2b8a21da57

How Calibration Content Shapes Attention-Based Reranking — arXiv:2609.17764

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.17764
License (as archived)
CC0-1.0
Archived
2026-10-15T13:49:15.233Z
SHA-256 of the archived text
837d3aada1ab0b6c04e3cb38e68a08e729fa2930a1cd741bd6005a77e0e2e5e5
Stable URL of this page
https://metraipolis.com/documento/dc_ff2b8a21da57

How to cite

How Calibration Content Shapes Attention-Based Reranking — arXiv:2609.17764 — arXiv (en), https://arxiv.org/abs/2609.17764 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff2b8a21da57 (Metraipolis archive dc_ff2b8a21da57, sha256 837d3aada1ab0b6c04e3cb38e68a08e729fa2930a1cd741bd6005a77e0e2e5e5)
BibTeX
@misc{dcff2b8a21da57,
  title = {How Calibration Content Shapes Attention-Based Reranking — arXiv:2609.17764},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17764},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff2b8a21da57; sha256 837d3aada1ab0b6c04e3cb38e68a08e729fa2930a1cd741bd6005a77e0e2e5e5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17764. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Attention-based rerankers score documents by aggregating query-to-document attention and subtracting a null-query calibration pass to remove positional and structural bias. Although widely used, this calibration assumes that the null pass removes irrelevant signal from each document. We show that modern prompt content, e.g. constraints, instructions, personas, and demonstrations can violate this assumption when it enters the scoring readout, making the null pass relevance-aware rather than null. We find that calibration is especially harmful when applied to prompts containing longer, more detailed instructions as the null-pass step removes relevant signal. Based on these findings, we propose interpolated null calibration, a training-free modification that controls how much of the instruction content enters the null baseline. It recovers attention-based reranking performance on instruction-heavy tasks where standard calibration fails, while preserving calibration's benefits when the null pass remains relevance-agnostic. On instruction heavy tasks, the recovered rankings surpass generative rerankers. We also show that in-context demonstrations improve attention-based reranking with little calibration interference, since demonstrations act only through the query pass and leave the null pass unchanged.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2609.17764 — 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 2026-10-15T13:48:24.277Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).