Metraipolis › archive › dc_2542e6df58b8

Explore or Confirm? LLM Agents Misallocate Measurements Under Noise — arXiv:2610.25111

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.25111
License (as archived)
CC0-1.0
Archived
2026-10-20T14:58:52.059Z
SHA-256 of the archived text
971f8e76c2fa2142b9343643a3bdd51dc3bb2302190f8297de1c584d8f305683
Stable URL of this page
https://metraipolis.com/documento/dc_2542e6df58b8

How to cite

Explore or Confirm? LLM Agents Misallocate Measurements Under Noise — arXiv:2610.25111 — arXiv (en), https://arxiv.org/abs/2610.25111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2542e6df58b8 (Metraipolis archive dc_2542e6df58b8, sha256 971f8e76c2fa2142b9343643a3bdd51dc3bb2302190f8297de1c584d8f305683)
BibTeX
@misc{dc2542e6df58b8,
  title = {Explore or Confirm? LLM Agents Misallocate Measurements Under Noise — arXiv:2610.25111},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.25111},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2542e6df58b8; sha256 971f8e76c2fa2142b9343643a3bdd51dc3bb2302190f8297de1c584d8f305683},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.25111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As model reasoning gets cheaper, the physical experiment becomes the slow and costly step in discovery. Each experiment returns a noisy reading: a high value can mean a good candidate or a lucky draw. With a fixed budget, an AI agent must keep choosing between exploring a new candidate and confirming a promising one by re-testing it. We measure that choice in Wellocate, a simulated assay campaign with known true values. It is built on a generated landscape and on two real protein datasets, GB1 and PhoQ, scrambled so no model can answer from memory. We test five language models as minimal agents, with no tools and no memory beyond the history in each prompt. They misallocate in both directions. Three over-confirm, spending most of the budget re-testing. On the generated landscape one under-confirms, trusting single readings, and there every model ends with a worse final pick than a basic scripted strategy. Each failure has its own cause. The over-confirmers run short of new candidates worth trying: a list of untried candidates cuts their re-testing and, on real protein data, makes two of them pick much better. A running summary of their readings or a description of the landscape changes it by at most five percentage points. For the under-confirmer, re-testing its top candidates in a final round, replayed on its own campaigns, recovers only about a quarter of what it loses by picking the wrong candidate. This suggests confirmation has to run through the whole campaign. We release Wellocate as a benchmark and reinforcement-learning environment with frozen splits and scripted baselines. With it, anyone can measure how an agent splits its budget before trusting it with real experiments, and train agents to split it better. Code, data and the environment are available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.25111 — 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 2026-10-20T14:58:50.770Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).