Metraipolis › archive › dc_df3530ea03e2

When May an Agent Stop? Evidence-Carrying Termination for Tool-Using LLMs — arXiv:2608.23623

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.23623
License (as archived): CC0-1.0
Archived: 2026-10-18T06:36:30.779Z
SHA-256 of the archived text: 16e5adfdebc583833dad2f41f96cba151149849c50f326169ac1dad1e078d85d
Stable URL of this page: https://metraipolis.com/documento/dc_df3530ea03e2

How to cite

When May an Agent Stop? Evidence-Carrying Termination for Tool-Using LLMs — arXiv:2608.23623 — arXiv (en), https://arxiv.org/abs/2608.23623 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_df3530ea03e2 (Metraipolis archive dc_df3530ea03e2, sha256 16e5adfdebc583833dad2f41f96cba151149849c50f326169ac1dad1e078d85d)

BibTeX

@misc{dcdf3530ea03e2,
  title = {When May an Agent Stop? Evidence-Carrying Termination for Tool-Using LLMs — arXiv:2608.23623},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.23623},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_df3530ea03e2; sha256 16e5adfdebc583833dad2f41f96cba151149849c50f326169ac1dad1e078d85d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.23623. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Tool-using agents must decide when to stop. Existing systems already gate terminal success, certify execution traces, or enforce runtime polici es, but do not test this particular receipt-, scope-, and closed-replay design at the COMPLETE boundary across controlled termination faults. W e instantiate and evaluate Evidence-Carrying Termination (ECT): an agent may return COMPLETE only when a typed certificate binds every required answer claim to valid, in-scope trace evidence and a deterministic replay reconstructs the claimed value. A locked static study crosses 48 ful ly synthetic tasks in six tool-use families with clean execution and eight faults. ECT produced 0/288 unsafe completions versus 252/288 for the inspected termination-critic core (difference -87.50 pp, 95% task-cluster interval [-87.50, -87.50] pp). A fresh, prespecified and frozen 576- trajectory study then compares ECT with the critic core, its faithful controller, and a full-trace LLM critic. On 22 primary held-out task clus ters, ECT produced 0/66 premature unsupported terminations versus 40/66 for the controller (difference -60.61 pp, 95% interval [-78.79, -40.91] pp), while supported completion was 97/132 versus 92/132 (difference 3.79 pp, interval [0.00, 9.09] pp), satisfying a -10-point noninferiority margin. ECT executed successful recovery in 18/66 trajectories, of which 17 subsequently completed with support; all three closed-loop gates p assed. ECT certifies support in a recorded trace under declared assumptions, not external truth, safety, or alignment.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-26 in cs.LG).
Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.23623 — 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 2026-10-18T06:28:52.298Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-26?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).