Metraipolis › archive › dc_61625d81eef8

Schema: Discovering Unknown Environments via Agentic Program Induction — arXiv:2609.39140

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39140
License (as archived)
CC0-1.0
Archived
2026-10-07T12:42:31.361Z
SHA-256 of the archived text
a9edfcccdbf10705976b191c0ccbde953ca6c440b55707c02ed0cdb9e195ec46
Stable URL of this page
https://metraipolis.com/documento/dc_61625d81eef8

How to cite

Schema: Discovering Unknown Environments via Agentic Program Induction — arXiv:2609.39140 — arXiv (en), https://arxiv.org/abs/2609.39140 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_61625d81eef8 (Metraipolis archive dc_61625d81eef8, sha256 a9edfcccdbf10705976b191c0ccbde953ca6c440b55707c02ed0cdb9e195ec46)
BibTeX
@misc{dc61625d81eef8,
  title = {Schema: Discovering Unknown Environments via Agentic Program Induction — arXiv:2609.39140},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39140},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_61625d81eef8; sha256 a9edfcccdbf10705976b191c0ccbde953ca6c440b55707c02ed0cdb9e195ec46},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39140. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learning to complete tasks in unfamiliar environments with unknown rules remains a key challenge for LLM agents. Current LLM agents often record their discoveries in prose, which may not provide a compact, explicit account of how the environment works. Inspired by how scientists organize observations into testable, predictive theories, we introduce Schema, an agent harness that organizes learning and action through interactive program induction. The LLM agent decides what to investigate and how to act, expressing its evolving understanding of the environment as executable programs. The harness consists of a persistent program workspace and a small set of interfaces for checking these programs against the interaction history, planning within them, and executing plans under step-by-step verification. Schema raises ARC-AGI-3 RHAE from 58.7% to 99.2% with the same base model, solves 100% of the public DiG-bench games, and reaches the median performance of the top-50 human players on MazeBench. Extensive analysis shows the effectiveness of Schema in unknown mechanism discovery, and ablations confirm the contribution of each component.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.39140 — 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 2026-10-07T12:14:13.735Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).