Metraipolis › archive › dc_857e5400a34a

PACE: Research Proving Agent for Adaptive Conjecture Exploration — arXiv:2610.33690

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.33690
License (as archived): CC0-1.0
Archived: 2026-10-27T07:57:18.648Z
SHA-256 of the archived text: 893ec518bca12b473ba689f70d2302c08dc19a814e415f44a00a660921dbf7f4
Stable URL of this page: https://metraipolis.com/documento/dc_857e5400a34a

How to cite

PACE: Research Proving Agent for Adaptive Conjecture Exploration — arXiv:2610.33690 — arXiv (en), https://arxiv.org/abs/2610.33690 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_857e5400a34a (Metraipolis archive dc_857e5400a34a, sha256 893ec518bca12b473ba689f70d2302c08dc19a814e415f44a00a660921dbf7f4)

BibTeX

@misc{dc857e5400a34a,
  title = {PACE: Research Proving Agent for Adaptive Conjecture Exploration — arXiv:2610.33690},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33690},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_857e5400a34a; sha256 893ec518bca12b473ba689f70d2302c08dc19a814e415f44a00a660921dbf7f4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33690. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Many mathematical research agents focus on solving problems with fixed statements. However, conjecture research, especially in applied mathematics, may require revising candidate questions or conclusions as investigation proceeds, coupling problem formulation with proof development. We present PACE, a research-proof agent that supports this process through a structured research loop that selects mathematical objectives and develops or challenges candidate mechanisms. The whole prover constructs proofs from these mechanisms, and an independent verifier checks them against the original target. Evidence-tracking memory stores assumptions and dependencies arising during research and marks evidence for revalidation when a candidate or condition changes. This memory supports task-specific context composition that keeps checked premises, exploratory material and unresolved obligations distinct. To evaluate proof construction and conjecture research, we introduce ResearchProofBench, an applied-mathematics benchmark focused on optimization, comprising 51 known-proof tasks with 331 hidden checkpoints under historical abstract access and six open-problem cases. The known-proof tasks assess reconstruction of published results, while the open-problem cases assess mathematical investigation and proof development. On the 51 known-proof tasks, PACE achieves an aggregate Proof Score of 75.16 and completes 20 proofs. In conjecture research, PACE develops conditional results, with its submissions receiving higher review scores than the compared baselines.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.33690 — 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 2026-10-27T07:55:30.660Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).