Puls Dane publiczne, aktualizowane automatycznie
arXiv – nowe publikacje publikacja

Cogentic: Multi-Agent Orchestration for Automated Proof Discovery

[Sztuczna inteligencja] Autorzy: Yang Cai, Vineet Gupta, Yanchen Jiang i 4 więcej

We present Cogentic, a multi-agent harness for automated proof discovery on open research problems. While frontier language models can generate strong mathematical ideas in a single shot, single-shot generation is often insufficient for open problems that require exploring multiple competing conjectures, overcoming subtle technical obstructions, and retaining intermediate progress over a long horizon. Cogentic addresses these challenges through an iterative prove--verify loop in which an orchestrator allocates a population of independent provers across distinct proof directions, subjects their output to adversarial verification by several specialized components, and promotes confirmed intermediate results into a persistent verified ledger that later rounds build on. The harness is designed to be able to solve research-level math and theoretical computer science problems. Using Gemini as…

Kategoria arxiv
cs.AI
Kategoria
Sztuczna inteligencja
Pdf
https://arxiv.org/pdf/2609.40324v1

Otwórz oryginał w źródle ↗

Więcej z tego źródła

Wszystkie