Metraipolis › archive › dc_c700317dd608
L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents — arXiv:2610.19591
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19591
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T14:23:29.768Z
- SHA-256 of the archived text
1dc501ccd2dbbd89ba3653cbd931ae04d0d4dedde01896e231734e996964ea9c- Stable URL of this page
https://metraipolis.com/documento/dc_c700317dd608
How to cite
L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents — arXiv:2610.19591 — arXiv (en), https://arxiv.org/abs/2610.19591 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c700317dd608 (Metraipolis archive dc_c700317dd608, sha256 1dc501ccd2dbbd89ba3653cbd931ae04d0d4dedde01896e231734e996964ea9c)
BibTeX
@misc{dcc700317dd608,
title = {L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents — arXiv:2610.19591},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19591},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c700317dd608; sha256 1dc501ccd2dbbd89ba3653cbd931ae04d0d4dedde01896e231734e996964ea9c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19591. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
The lasting gains in AI have come from methods that absorb more computation, and language models already scale computation within a call. Useful work, though, happens across calls: over hours or days, through tools and external systems, among many agents and the humans they work for, and increasingly through code the agents write themselves. Whether that work compounds is enabled by the harness around it, and the harness faces a constraint the model does not: of the three participants in such a computation, the human, the model and code, only the human's attention does not scale, so the computation can grow only if the human is not in every loop. L3M is a runtime built on that premise, written in Lean 4 so that one language implements the harness, states its invariants, and is the language agents extend it in. One computation spans humans, agents, models and code, and all of its state lives in Git: a rich state trajectory -- rather than a chat conversation -- is an object the runtime can resume, rewind, fork, compare and merge. Waiting, communication, delegation and undo are primitives, not patterns the model reasons through each time. When an agent works out a procedure, it can write it once as a sealed Lean program whose type is its permission request, and the harness offers it back as a tool -- machinery for the model to use. To take the human out of the loop, L3M uses reversibility to decide where that is safe: a tool's review tier follows from how far its effects can be undone, so agents act freely where the runtime can recover. Because the machinery can change, the properties that make this safe are stated as kernel-checked theorems about the code that runs. On Terminal-Bench, an untuned L3M reaches about two thirds of its (unsafe) shell-equipped pass rate using only safe (not needing human review) tools. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.19591 — 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 2026-10-15T13:53:35.882Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).