arXiv：cs.AI· Md Asiful Islam, Fahmida Alam, Mihai Surdeanu·· 16 小时前
LR-Guard：通过潜在双假设推理与按需轨迹重建实现高效安全护栏
Efficient Safety Guardrails through Latent Dual-Hypothesis Reasoning with On-Demand Trace Reconstruction
AI 导读
LR-Guard 是一种高效可解释的安全护栏，通过将冻结教师模型对安全与不安全提示词解析的分析蒸馏为紧凑潜在表示，同时学习结构化潜在推理与按需推理重建，无需为每次决策生成中间文本即可提供人类可读解释。
正文
Abstract:Guardrail approaches that rely on reasoning-based large language models (LLM) improve safety decisions, but come with substantial inference overhead due to the explicit reasoning. We introduce LR-Guard (Latent Reasoning Guardrail), an efficient and explainable guardrail that jointly learns structured latent reasoning and on-demand reasoning reconstruction. Our method distills a frozen teacher's analysis of safe and unsafe prompt interpretations and their comparison into compact latent representations. Joint training optimizes these representations for both safety prediction and reconstruction of the teacher's explicit reasoning traces, enabling human-readable explanations without requiring intermediate text generation for every decision. Experiments show that LR-Guard outperforms its explicit-reasoning teacher in most in-domain and out-of-domain settings on safety classification and explanation extraction, while achieving a 19.4x inference speedup at batch size one. Further, reconstructed reasoning traces from the latent information exhibit high semantic similarity to the teacher's explicit reasoning.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.18304 [cs.AI] |
| (or arXiv:2610.18304v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.18304 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Md Asiful Islam [view email]
[v1]
Tue, 13 Oct 2026 12:49:47 UTC (394 KB)
来源：arXiv：cs.AI · arxiv.org