Metraipolis › archive › dc_a00809d481c3
Satisfaction Is Not Explanation: Auditing Vacuity and Training Influence in Temporal-Logic-Guided Reinforcement Learning — arXiv:2609.27743
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27743
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:06:28.558Z
- SHA-256 of the archived text
ae209d419d4ec8ceadc7c61c9d5231ae1400a2a9194555b5920cb9d8cc180ddd- Stable URL of this page
https://metraipolis.com/documento/dc_a00809d481c3
How to cite
Satisfaction Is Not Explanation: Auditing Vacuity and Training Influence in Temporal-Logic-Guided Reinforcement Learning — arXiv:2609.27743 — arXiv (en), https://arxiv.org/abs/2609.27743 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a00809d481c3 (Metraipolis archive dc_a00809d481c3, sha256 ae209d419d4ec8ceadc7c61c9d5231ae1400a2a9194555b5920cb9d8cc180ddd)
BibTeX
@misc{dca00809d481c3,
title = {Satisfaction Is Not Explanation: Auditing Vacuity and Training Influence in Temporal-Logic-Guided Reinforcement Learning — arXiv:2609.27743},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27743},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a00809d481c3; sha256 ae209d419d4ec8ceadc7c61c9d5231ae1400a2a9194555b5920cb9d8cc180ddd},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27743. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A reinforcement learning policy that satisfies its temporal-logic specification has passed a test, not an assurance argument. The clause that matters to a reviewer may never have mattered to training: it may have been avoided entirely, forced by the environment regardless of what the policy learned, or redundant next to the ordinary task reward. Satisfaction probability and task return cannot tell any of this apart. This paper introduces an audit layer that can. It measures whether a specification clause was actually exercised, whether that role was forced or chosen, and whether the obvious way to test causation, weakening the clause and retraining, is even valid. Often it is not: we prove that comparable weaker/stronger training objectives can share perfect optima under standard acceptance-derived rewards, show related ablation hazards across a large corpus of published specifications, and then show that a properly designed intervention detects the effect it should. Across standard reinforcement learning benchmarks and published external artifacts, the audit layer separates six regimes that a single satisfaction number collapses into one. A policy that satisfies its specification has answered whether. This paper asks why. Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-24 in cs.SE). Categorie arXiv: Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2609.27743 — 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 2026-10-08T10:13:35.115Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).