对我的谎言：推理模型中的思维链推理可信度如何？
Lie to Me: How Faithful Is Chain-of-Thought Reasoning in Reasoning Models?
AI总结：
研究评估了12种开源推理模型在MMLU和GPQA Diamond中的表现，发现模型在提示影响下的可信度差异显著，训练方法和架构影响更大。
摘要：
链式思维（CoT）推理已被提出作为大型语言模型在安全关键部署中的透明机制，但其有效性取决于忠实性（模型是否准确描述影响其输出的因素）。此前仅在两个专有模型上进行了评估，发现Claude 3.7 Sonnet的承认率低至25%，DeepSeek-R1为39%。为扩展此评估至开源领域，本研究测试了12种开源推理模型，涵盖9种架构家族（7B-685B参数），在498个多选问题上注入了六类推理提示（奉承、一致性、视觉模式、元数据、黑客破解和不道德信息），并测量模型在提示成功改变答案时承认提示影响的比率。在41,832次推理运行中，整体忠实率从39.7%（Seed-1.6-Flash）到89.9%（DeepSeek-V3.2-Speciale）不等，其中一致性提示（35.5%）和奉承提示（53.9%）表现出最低的承认率。训练方法和模型家族比参数数量更能预测忠实性，关键词分析揭示了思考令牌承认（约87.5%）与答案文本承认（约28.6%）之间的显著差距，表明模型内部识别提示影响，但系统性地在输出中抑制这种承认。这些发现对CoT监控作为安全机制的可行性有直接影响，并表明忠实性不是推理模型的固定属性，而是受架构、训练方法和影响提示性质系统性影响。
英文摘要：
Chain-of-thought (CoT) reasoning has been proposed as a transparency mechanism for large language models in safety-critical deployments, yet its effectiveness depends on faithfulness (whether models accurately verbalize the factors that actually influence their outputs), a property that prior evaluations have examined in only two proprietary models, finding acknowledgment rates as low as 25% for Claude 3.7 Sonnet and 39% for DeepSeek-R1. To extend this evaluation across the open-weight ecosystem, this study tests 12 open-weight reasoning models spanning 9 architectural families (7B-685B parameters) on 498 multiple-choice questions from MMLU and GPQA Diamond, injecting six categories of reasoning hints (sycophancy, consistency, visual pattern, metadata, grader hacking, and unethical information) and measuring the rate at which models acknowledge hint influence in their CoT when hints successfully alter answers. Across 41,832 inference runs, overall faithfulness rates range from 39.7% (Seed-1.6-Flash) to 89.9% (DeepSeek-V3.2-Speciale) across model families, with consistency hints (35.5%) and sycophancy hints (53.9%) exhibiting the lowest acknowledgment rates. Training methodology and model family predict faithfulness more strongly than parameter count, and keyword-based analysis reveals a striking gap between thinking-token acknowledgment (approximately 87.5%) and answer-text acknowledgment (approximately 28.6%), suggesting that models internally recognize hint influence but systematically suppress this acknowledgment in their outputs. These findings carry direct implications for the viability of CoT monitoring as a safety mechanism and suggest that faithfulness is not a fixed property of reasoning models but varies systematically with architecture, training method, and the nature of the influencing cue.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
