框架对大语言模型伦理立场的不稳定性影响：道德困境中否定敏感性的审计
Framing Instability in LLM Ethical Stance: Auditing Negation Sensitivity in Moral Dilemmas
AI总结：
研究大语言模型在伦理问题上因框架改变导致立场不稳定的问题，通过极性配对提议审计16个模型，发现小型模型波动大，商业模型也有变化，提出否定敏感性指数来衡量立场稳定性，指出立场易翻转的模型在高风险决策中不可靠。
摘要：
语言模型越来越多地被用于处理具有伦理后果的问题，但模型所表达的立场可能会因框架的改变而变化。我们使用极性配对提议（“他们应该X”/“他们不应该X”）对16个模型在14个充满伦理困境的问题上进行了审计。一个模型对潜在行为的判断不应仅仅因为问题被表述为禁止而非规定而反转，但我们发现存在系统性偏差，包括完全的支持翻转，这表明伦理决策容易受到框架不稳定性的影响。小型开放权重模型（1 - 4B参数）在肯定框架下对提议行为的支持率为24%，但在否定框架下高达100%，波动高达76个百分点。对回应样本的人工编码证实了这种不稳定性是真实的，同时表明二元同意/不同意代理夸大了其程度，这意味着大语言模型法官不能取代人类编码员，因为它默默地合并了弃权情况并反映了正在研究的强制选择偏差。商业模型大多更稳定，但仍有显著变化，跨模型一致性从单纯肯定框架下的73%降至简单否定框架下的59%。我们认为，由于二元同意/不同意格式既夸大了表面支持又掩盖了极性依赖性，单句式审计可能会错误报告模型的伦理立场，因此我们提出否定敏感性指数（NSI）作为一种直接测量立场稳定性的补充方法。一个立场随措辞翻转的模型在任何高风险决策场景中都不可靠。
英文摘要：
Language models are increasingly consulted on ethically consequential questions, yet the stance a model expresses may not survive a change in framing. We audit 16 models across 14 ethically fraught dilemmas using polarity-paired proposals ("They should X" / "They should not X"). A model's judgment of the underlying action should not reverse merely because the question is phrased as a prohibition rather than a prescription and yet, we find systematic deviations from this invariance including wholesale endorsement flips, indicating that ethical decisions are vulnerable to framing instability. Small open-weight models (1-4B parameters) endorse a proposed action 24% of the time under affirmative framing but up to 100% under negated framings, a swing of as much as 76 percentage points. Human coding of a response sample confirms the instability is genuine while showing that binary agree/disagree proxies over-state its magnitude, suggesting that an LLM judge cannot replace human coders because it silently collapses abstentions and mirrors the very forced-choice bias under study. Commercial models are for the most part more stable but still shift substantially, with cross-model agreement dropping from 73% on the bare affirmative framing to 59% under simple negation. We argue that because binary agree/disagree formats both inflate apparent endorsement and mask polarity-dependence, single-phrasing audits can misreport a model's ethical stance, and we propose the Negation Sensitivity Index (NSI) as a complement that measures stance stability directly. A model whose stance flips with phrasing cannot be relied upon in any high-stakes decision scenario.
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。
