Metraipolis › archive › dc_22feb4e306f9

Search When Needed, Proceed When Reliable: Metacognitive Reliability Supervision for Efficient Agentic Search — arXiv:2610.41738

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41738
License (as archived): CC0-1.0
Archived: 2026-11-02T05:02:52.218Z
SHA-256 of the archived text: 05403e2fe16f4ea42afe00033d0ecdaa3a86ca97727d220f76e0aadef34bae9e
Stable URL of this page: https://metraipolis.com/documento/dc_22feb4e306f9

How to cite

Search When Needed, Proceed When Reliable: Metacognitive Reliability Supervision for Efficient Agentic Search — arXiv:2610.41738 — arXiv (en), https://arxiv.org/abs/2610.41738 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_22feb4e306f9 (Metraipolis archive dc_22feb4e306f9, sha256 05403e2fe16f4ea42afe00033d0ecdaa3a86ca97727d220f76e0aadef34bae9e)

BibTeX

@misc{dc22feb4e306f9,
  title = {Search When Needed, Proceed When Reliable: Metacognitive Reliability Supervision for Efficient Agentic Search — arXiv:2610.41738},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41738},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_22feb4e306f9; sha256 05403e2fe16f4ea42afe00033d0ecdaa3a86ca97727d220f76e0aadef34bae9e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41738. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Efficient agentic search requires large language models (LLMs) to determine whether they can reliably proceed with their current knowledge or need additional external evidence. Supervision based on final-answer correctness or trajectory-level feedback evaluates realized trajectory performance. However, observed trajectory performance is jointly shaped by stochastic generation, retrieval quality, and downstream reasoning. Such signals therefore provide only indirect evidence of how reliable the current decision state was and how retrieved evidence changed its reliability. An effective reward should encourage search when external information is needed and favor searches that contribute to the answer. To this end, we propose a metacognitive framework for training LLMs to make search decisions based on the reliability of their intermediate reasoning states. Our framework estimates the reliability of local conclusions to distinguish whether the current information supports proceeding without search from how much retrieved evidence improves that reliability. The current-state reliability score assesses whether the model can proceed without search. The search-action score measures how retrieved evidence changes estimated reliability. Together, these scores define a reliability-aware process reward for Group Relative Policy Optimization (GRPO), providing finer-grained supervision for search and no-search behavior. Experiments across seven question-answering benchmarks demonstrate that our method achieves a better accuracy-efficiency trade-off than strong agentic-search baselines. Our code and model checkpoints will be open-sourced.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.IR, nuovi lavori del giorno).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2610.41738 — 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 2026-11-02T04:48:58.167Z dalla pagina https://arxiv.org/list/cs.IR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).