Metraipolis › archive › dc_6dcc31bf7716
Commitment Hierarchies under Intent Revision: A Belief-Revision Account of Salvage in Tool-Use Agents — arXiv:2609.37453
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37453
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:23:53.508Z
- SHA-256 of the archived text
1238f7d10905c718d88a077997c6aa3b04b14c7dcb1d78fc6f4ed872a0cf72d2- Stable URL of this page
https://metraipolis.com/documento/dc_6dcc31bf7716
How to cite
Commitment Hierarchies under Intent Revision: A Belief-Revision Account of Salvage in Tool-Use Agents — arXiv:2609.37453 — arXiv (en), https://arxiv.org/abs/2609.37453 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6dcc31bf7716 (Metraipolis archive dc_6dcc31bf7716, sha256 1238f7d10905c718d88a077997c6aa3b04b14c7dcb1d78fc6f4ed872a0cf72d2)
BibTeX
@misc{dc6dcc31bf7716,
title = {Commitment Hierarchies under Intent Revision: A Belief-Revision Account of Salvage in Tool-Use Agents — arXiv:2609.37453},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37453},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6dcc31bf7716; sha256 1238f7d10905c718d88a077997c6aa3b04b14c7dcb1d78fc6f4ed872a0cf72d2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37453. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
When a user changes their mind partway through a task, an agent that has already split the task into sub-goals and paid for tool calls must decide, per cached sub-result, whether to keep, patch, or discard it (salvage), restarting wastes valid work and continuing unchanged answers the old question. Our main finding is that salvage quality is a matter of role design rather than model capability: a language model asked the keep/patch/discard question one node at a time is unreliable, but asked to classify the revision once, with a deterministic layer propagating the decision, it reaches the cost-optimal oracle on all three models tested, from two vendors. Modeling the plan as a commitment hierarchy and the intent change as a belief-revision operator with AGM style postulates, we prove that no policy observing only a node's local view can be both safe and cost optimal, while the single classification design is both. Across three environments the policy recovers the full achievable savings, 43% cheaper than restart, at 100% correctness. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.37453 — 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 2026-10-07T15:54:53.628Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).