Metraipolis › archive › dc_5da53ea0be2c

SBCO: Self-Supervised, Verifier-Grounded Harness Optimization For Planning Agents — arXiv:2608.10157

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.10157
License (as archived): CC0-1.0
Archived: 2026-10-19T06:08:11.848Z
SHA-256 of the archived text: c12871c8eb7877b9d8e9d01f86b2e1c2fa44839cd150855d55b0a0c31a849fef
Stable URL of this page: https://metraipolis.com/documento/dc_5da53ea0be2c

How to cite

SBCO: Self-Supervised, Verifier-Grounded Harness Optimization For Planning Agents — arXiv:2608.10157 — arXiv (en), https://arxiv.org/abs/2608.10157 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5da53ea0be2c (Metraipolis archive dc_5da53ea0be2c, sha256 c12871c8eb7877b9d8e9d01f86b2e1c2fa44839cd150855d55b0a0c31a849fef)

BibTeX

@misc{dc5da53ea0be2c,
  title = {SBCO: Self-Supervised, Verifier-Grounded Harness Optimization For Planning Agents — arXiv:2608.10157},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10157},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5da53ea0be2c; sha256 c12871c8eb7877b9d8e9d01f86b2e1c2fa44839cd150855d55b0a0c31a849fef},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10157. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Self-improving agents seek to reduce the human engineering effort behind AI systems by enabling them to evolve and self-improve their performance over time. Recently, methods like the Darwin Gödel Machine and the Huxley Gödel Machine have been proposed which enable open-ended, recursive self-improvement through self-reference where a coding agent edits its own code. Such self-referential self-improvement methods require that the competence required to perform the task coincides or aligns well with the competence required for self-modification which is the case for coding tasks. For domains or tasks, which do not satisfy the alignment needed, self-referential self-improvement is not available. In such cases, it is possible to adapt the above algorithms to other tasks by removing the self-referential aspect or introducing explicit self-modification of a meta-agent -- both computationally expensive, relying on population or self-modification search over many candidate agents. For planning tasks with explicit constraints, we propose a far cheaper alternative. We introduce SBCO (Self-supervised Block Coordinate Optimizer), a verifier-grounded harness optimizer in the same closed-loop, improve-from-experience family as the Gödel-machine methods, but self-supervised rather than self-referential. Given an agentic harness, SBCO learns a decomposed bank of verifiers and a harness policy via approximate block coordinate ascent, improving the agent's outputs from its own graded feedback---with a fixed meta-agent and no human labels. Across two domains SBCO matches or exceeds a customized self-modifying baseline while using 4-5.5 times less compute budget.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.10157 — 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 2026-10-19T05:57:28.238Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).