{"ID":6023602,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T14:11:27.630055639Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.06326","arxiv_id":"2607.06326","title":"DT-Guard: Intent-Driven Reasoning-Active Training for Reasoning-Free LLM Safety Guardrail","abstract":"Large language models deployed in open-world applications require safety guardrails that are both robust to complex risks and efficient enough for low-latency runtime moderation. Existing guardrails face a practical trade-off between lightweight classification-based models, which are efficient but often struggle with concealed intent, ambiguous semantics, and borderline safety decisions, and reasoning-based guards, which improve judgment quality but introduce additional token generation and inference latency. We present DT-Guard, a content safety guardrail model based on a Reasoning-Active Training, Reasoning-Free Inference paradigm. The key idea is to use reasoning supervision during training while emitting only structured safety labels at inference time. DT-Guard formulates safety judgment as a progressive decision process, Intent - Category - Safety, and constructs an intent-driven dataset with intent labels, risk categories, safety labels, and structured reasoning trajectories. To further improve hard-case robustness, we propose Rollout-Guided Progressive Hard-Case Optimization (RG-PHO), which uses multi-rollout consistency to identify stably mastered, persistently failed, and preference-unstable samples, and applies targeted supervised and preference optimization accordingly. At inference time, DT-Guard directly generates structured labels without explicit reasoning traces, preserving deployment efficiency. Experiments on prompt-side and response-side safety benchmarks show that DT-Guard achieves average F1 scores of 0.886 and 0.870, respectively. With only a 4B backbone, it reaches a dual-side average F1 of 0.878, outperforming strong 8B guardrail baselines. These results demonstrate that reasoning supervision can be effectively internalized into low-latency safety discrimination.","short_abstract":"Large language models deployed in open-world applications require safety guardrails that are both robust to complex risks and efficient enough for low-latency runtime moderation. Existing guardrails face a practical trade-off between lightweight classification-based models, which are efficient but often struggle with c...","url_abs":"https://arxiv.org/abs/2607.06326","url_pdf":"https://arxiv.org/pdf/2607.06326v1","authors":"[\"He Liu\",\"Changtao Miao\",\"Xinjie Yang\",\"Tianle Song\",\"Yin Wu\",\"Junchi Chen\",\"Bintao He\",\"Xinyuan Zhang\",\"Bo Zhang\",\"Shi Yan\",\"Wei Lu\",\"Wei Wang\",\"Danyang Xu\",\"Jiansheng Cai\",\"Zhe Li\"]","published":"2026-07-07T14:25:05Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}