Metraipolis › archive › dc_cdbd5b7fe853

CASI: Credit Assignment via Set Intervention for Agentic Retrieval — arXiv:2610.20293

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20293
License (as archived)
CC0-1.0
Archived
2026-10-15T14:23:48.499Z
SHA-256 of the archived text
a8ca21dbf8822c70421aef93f2add186037acc85949a23271108f1e1c6d7353b
Stable URL of this page
https://metraipolis.com/documento/dc_cdbd5b7fe853

How to cite

CASI: Credit Assignment via Set Intervention for Agentic Retrieval — arXiv:2610.20293 — arXiv (en), https://arxiv.org/abs/2610.20293 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cdbd5b7fe853 (Metraipolis archive dc_cdbd5b7fe853, sha256 a8ca21dbf8822c70421aef93f2add186037acc85949a23271108f1e1c6d7353b)
BibTeX
@misc{dccdbd5b7fe853,
  title = {CASI: Credit Assignment via Set Intervention for Agentic Retrieval — arXiv:2610.20293},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20293},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cdbd5b7fe853; sha256 a8ca21dbf8822c70421aef93f2add186037acc85949a23271108f1e1c6d7353b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20293. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models increasingly generate multiple retrieval actions jointly, while downstream systems often provide only one reward for the resulting set. Broadcasting this reward provides a valid score-function gradient estimator for the expected set-level objective, but does not determine a coalition-dependent within-set attribution. We introduce CASI (Credit Assignment via Set Intervention for Agentic Retrieval), which constructs the missing counterfactual information through snapshot-grounded set intervention: generate and retrieve once, hold replayable snapshot components fixed, and evaluate alternative subsets under common conditions. Shapley aggregation converts these marginals into within-set credit, which bounded zero-sum advantage shaping incorporates into policy optimization. For m >= 4, singleton, leave-one-out, and full-set endpoints do not generally determine the Shapley allocation; under additive set values, extra coalition evaluations are redundant. In a five-action sponsored-search setting, exhaustive auction replay over 1,200 requests reveals substantial coalition dependence. Under the training-time evaluator, CASI achieves the highest median cosine similarity to a replay-defined Shapley reference among evaluated credit rules; under a sensitivity backend, CASI does not outperform Singleton. Across five independent training runs, CASI-trained policies achieve higher held-out auction-replay set value than Singleton- or Broadcast-trained policies. In a randomized seven-day production A/B test on 10% of traffic, the CASI-trained policy yields an observed 3.51% lift in the primary advertiser-value metric over production control.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Information Retrieval (cs.IR); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.20293 — 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 2026-10-15T13:53:35.949Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).