Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.15588 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915746707144704 |
|---|---|
| author | Lin, Junyu Liu, Meizhen Huang, Xiufeng Li, Jinfeng Hong, Haiwen Yuan, Xiaohan Chen, Yuefeng Huang, Longtao Xue, Hui Duan, Ranjie Chen, Zhikai Fu, Yuchuan Li, Defeng Gao, Lingyao Yang, Yitong |
| author_facet | Lin, Junyu Liu, Meizhen Huang, Xiufeng Li, Jinfeng Hong, Haiwen Yuan, Xiaohan Chen, Yuefeng Huang, Longtao Xue, Hui Duan, Ranjie Chen, Zhikai Fu, Yuchuan Li, Defeng Gao, Lingyao Yang, Yitong |
| contents | As large language models (LLMs) are increasingly deployed in real-world applications, safety guardrails are required to go beyond coarse-grained filtering and support fine-grained, interpretable, and adaptable risk assessment. However, existing solutions often rely on rapid classification schemes or post-hoc rules, resulting in limited transparency, inflexible policies, or prohibitive inference costs. To this end, we present YuFeng-XGuard, a reasoning-centric guardrail model family designed to perform multi-dimensional risk perception for LLM interactions. Instead of producing opaque binary judgments, YuFeng-XGuard generates structured risk predictions, including explicit risk categories and configurable confidence scores, accompanied by natural language explanations that expose the underlying reasoning process. This formulation enables safety decisions that are both actionable and interpretable. To balance decision latency and explanatory depth, we adopt a tiered inference paradigm that performs an initial risk decision based on the first decoded token, while preserving ondemand explanatory reasoning when required. In addition, we introduce a dynamic policy mechanism that decouples risk perception from policy enforcement, allowing safety policies to be adjusted without model retraining. Extensive experiments on a diverse set of public safety benchmarks demonstrate that YuFeng-XGuard achieves stateof-the-art performance while maintaining strong efficiency-efficacy trade-offs. We release YuFeng-XGuard as an open model family, including both a full-capacity variant and a lightweight version, to support a wide range of deployment scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15588 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | YuFeng-XGuard: A Reasoning-Centric, Interpretable, and Flexible Guardrail Model for Large Language Models Lin, Junyu Liu, Meizhen Huang, Xiufeng Li, Jinfeng Hong, Haiwen Yuan, Xiaohan Chen, Yuefeng Huang, Longtao Xue, Hui Duan, Ranjie Chen, Zhikai Fu, Yuchuan Li, Defeng Gao, Lingyao Yang, Yitong Computation and Language As large language models (LLMs) are increasingly deployed in real-world applications, safety guardrails are required to go beyond coarse-grained filtering and support fine-grained, interpretable, and adaptable risk assessment. However, existing solutions often rely on rapid classification schemes or post-hoc rules, resulting in limited transparency, inflexible policies, or prohibitive inference costs. To this end, we present YuFeng-XGuard, a reasoning-centric guardrail model family designed to perform multi-dimensional risk perception for LLM interactions. Instead of producing opaque binary judgments, YuFeng-XGuard generates structured risk predictions, including explicit risk categories and configurable confidence scores, accompanied by natural language explanations that expose the underlying reasoning process. This formulation enables safety decisions that are both actionable and interpretable. To balance decision latency and explanatory depth, we adopt a tiered inference paradigm that performs an initial risk decision based on the first decoded token, while preserving ondemand explanatory reasoning when required. In addition, we introduce a dynamic policy mechanism that decouples risk perception from policy enforcement, allowing safety policies to be adjusted without model retraining. Extensive experiments on a diverse set of public safety benchmarks demonstrate that YuFeng-XGuard achieves stateof-the-art performance while maintaining strong efficiency-efficacy trade-offs. We release YuFeng-XGuard as an open model family, including both a full-capacity variant and a lightweight version, to support a wide range of deployment scenarios. |
| title | YuFeng-XGuard: A Reasoning-Centric, Interpretable, and Flexible Guardrail Model for Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.15588 |