Metraipolis › archive › dc_52177aafa248
AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails — arXiv:2610.08923
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.08923
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:06:43.009Z
- SHA-256 of the archived text
8582ce82348bf9a4994ce239ab68c5f937ac2850baa5d4c4cae037e766c1a4d4- Stable URL of this page
https://metraipolis.com/documento/dc_52177aafa248
How to cite
AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails — arXiv:2610.08923 — arXiv (en), https://arxiv.org/abs/2610.08923 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_52177aafa248 (Metraipolis archive dc_52177aafa248, sha256 8582ce82348bf9a4994ce239ab68c5f937ac2850baa5d4c4cae037e766c1a4d4)
BibTeX
@misc{dc52177aafa248,
title = {AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails — arXiv:2610.08923},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.08923},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_52177aafa248; sha256 8582ce82348bf9a4994ce239ab68c5f937ac2850baa5d4c4cae037e766c1a4d4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08923. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Enterprise generative AI applications require robust safety mechanisms that can accommodate diverse risk postures, evolving policies, and varying latency constraints. Current guardrail solutions often suffer from rigidity, relying on fixed policy sets and offering limited transparency or reasoning flexibility. We present Adaguard, an adaptive LLM-as-a-Judge framework designed to address these challenges through dynamic policy enforcement and adaptive reasoning-budget allocation. Built using supervised fine-tuning (SFT) and reinforcement learning (GRPO), AdaGuard generalizes to user-defined safety and compliance policies at runtime without requiring frequent model updates. A core innovation of our approach is the ability to dynamically infer the complexity of input-policy pairs, allowing the model to switch between high-speed black-box inference and explainable, reasoning-enabled moderation. This flexibility enables developers to balance stringent latency requirements with the need for actionable transparency. This adaptive capability allows AdaGuard to rival other guardrail and frontier models several times its size, while its auto-reasoning mode recovers the accuracy of always-on reasoning at a fraction of the latency Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.08923 — 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 2026-10-08T10:14:05.881Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).