Metraipolis › archive › dc_3fc38ae0ddd2

Historical-Feature Residual Adaptation for Sequential Screening — arXiv:2610.19315

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19315
License (as archived)
CC0-1.0
Archived
2026-10-15T14:15:24.242Z
SHA-256 of the archived text
1fb3ba59b056eb55fe56d8f5a841970729ede3ccc8fb05148608f4332c5e4c94
Stable URL of this page
https://metraipolis.com/documento/dc_3fc38ae0ddd2

How to cite

Historical-Feature Residual Adaptation for Sequential Screening — arXiv:2610.19315 — arXiv (en), https://arxiv.org/abs/2610.19315 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3fc38ae0ddd2 (Metraipolis archive dc_3fc38ae0ddd2, sha256 1fb3ba59b056eb55fe56d8f5a841970729ede3ccc8fb05148608f4332c5e4c94)
BibTeX
@misc{dc3fc38ae0ddd2,
  title = {Historical-Feature Residual Adaptation for Sequential Screening — arXiv:2610.19315},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19315},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3fc38ae0ddd2; sha256 1fb3ba59b056eb55fe56d8f5a841970729ede3ccc8fb05148608f4332c5e4c94},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19315. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sequential screening policies must use acquired feedback effectively under a fixed experiment budget. We study acquisition-time residual adaptation: a small ranking correction built from fixed historical features and anchored to base scores recorded before feedback is revealed. The interface accepts a history-conditioned scorer and a candidate feature table; its primary instantiation augments AssayFormer's ten-dimensional representation with 64 historical directions while retaining the pretrained model. In retrospective evaluation on 20 official-test screens from 15 recorded sources, a comparison fixed before target-outcome access increased source-macro fraction of hits recovered from 22.58% to 28.61%, exceeding all six prespecified comparators. A same-feature ordinary ranking head achieved 25.73%. Supplementary strong logistic, rank-fusion, and unprojected controls achieved 26.42%, 26.02%, and 26.78%, respectively. A separate model-independent pilot on 13 reused development screens reconstructs the features without AssayFormer assets. Across three seeds, residual adaptation improves random-forest hit recovery from 28.87% to 31.27%, while logistic recovery changes from 30.22% to 29.78%. The results show effective pretrained-policy adaptation and useful transfer to an independent nonlinear base, with gains that depend on the base model.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.19315 — 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 2026-10-15T13:53:35.568Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).