Metraipolis › archive › dc_3ef257fcdfae

SearchJev: A Fast and Calibrated System-1 Model for Search Agents — arXiv:2610.05107

Source: arXiv
Original: https://arxiv.org/abs/2610.05107
License (as archived): CC0-1.0
Archived: 2026-10-06T05:28:49.124Z
SHA-256 of the archived text: 7cbebc75a3387ad9ad6a6ac4bc24981704cb24a1307c5133471aee0a0234cd59
Stable URL of this page: https://metraipolis.com/documento/dc_3ef257fcdfae

How to cite

SearchJev: A Fast and Calibrated System-1 Model for Search Agents — arXiv:2610.05107 — arXiv, https://arxiv.org/abs/2610.05107 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3ef257fcdfae (Metraipolis archive dc_3ef257fcdfae, sha256 7cbebc75a3387ad9ad6a6ac4bc24981704cb24a1307c5133471aee0a0234cd59)

BibTeX

@misc{dc3ef257fcdfae,
  title = {SearchJev: A Fast and Calibrated System-1 Model for Search Agents — arXiv:2610.05107},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05107},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3ef257fcdfae; sha256 7cbebc75a3387ad9ad6a6ac4bc24981704cb24a1307c5133471aee0a0234cd59},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05107. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Search agents repeatedly make short decisions about relevance, evidence sufficiency, and search actions. Using generative language models for these decisions introduces latency and unreliable confidence. We present SearchJev, a fast and calibrated System-1 model that separates search decisions from System-2 reasoning and generation. Given a search state and a decision schema, SearchJev directly scores legal options without autoregressive output generation. We propose Soft-Label Learning for Calibrated Decisions (SLCD) to learn decision probabilities from uncertain supervision and calibrate their confidence. In a dual-system search agent, SearchJev handles short decisions and delegates uncertain judgments to System 2, which retains planning, query generation, and answer composition. We also introduce SearchDecision-Bench, a benchmark unifying six types of search decisions for training and evaluation. On SearchDecision-Bench, SEARCHJEV improves decision quality over same-size Qwen3.5 autoregressive models, achieves 5.2-5.3 times faster decisions, and reduces average expected calibration error by 41-74%. On BrowseComp-Plus, the dual-system agents achieve a 3.7-4.7 times speedup in active search time while improving answer accuracy from 45% to up to 54%.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Information Retrieval (cs.IR) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.05107 — 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 2026-10-06T05:24:20.430Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).