Large Reasoning Models (LRMs) have significantly improved problem-solving through explicit Chain-of-Thought (CoT) reasoning. However, this capability creates a Safety-Helpfulness Paradox: the reasoning process itself can be misused to justify harmful actions or conceal malicious intent behind lengthy intermediate steps. Most existing benchmarks only check the final output, missing how risks evolve, or ``drift'', during the model's internal reasoning. To address this, we propose SafeRBench, the first framework to evaluate LRM safety end-to-end, from the initial input to the reasoning trace and final answer. Our approach introduces: (i) a Risk Stratification Probing that uses specific risk levels to stress-test safety boundaries beyond simple topics; (ii) Micro-Thought Analysis, a new chunking method that segments traces to pinpoint exactly where safety alignment breaks down; and (iii) a comprehensive suite of 10 fine-grained metrics that, for the first time, jointly measure a model's Risk Exposure (e.g., risk level, execution feasibility) and Safety Awareness (e.g., intent awareness). Experiments on 19 LRMs reveal that while enabling Thinking modes improves safety in mid-sized models, it paradoxically increases actionable risks in larger models due to a strong always-help tendency.

翻译：大型推理模型（LRM）通过显式的思维链（CoT）推理显著提升了问题解决能力。然而，这种能力也引发了“安全-有益性悖论”：推理过程本身可能被滥用于为有害行为辩护，或将恶意意图隐藏在冗长的中间步骤之后。现有基准大多仅检查最终输出，未能捕捉模型内部推理过程中风险如何演变或“漂移”。为此，我们提出SafeRBench——首个对LRM安全性进行端到端评估的框架，涵盖从初始输入、推理轨迹到最终答案的全过程。我们的方法包含：（i）风险分层探测，利用特定风险等级对安全性边界进行压力测试，超越简单主题范畴；（ii）微思维分析，一种创新的分块方法，通过分割推理轨迹精确定位安全对齐失效的具体位置；（iii）一套包含10项细粒度指标的综合评估体系，首次联合衡量模型的风险暴露（如风险等级、执行可行性）与安全认知（如意图感知）。在19个LRM上的实验表明：虽然启用思考模式能提升中等规模模型的安全性，但由于强烈的“始终助人”倾向，该模式反而会增大大型模型的可操作风险。