压力测试模型规范揭示语言模型之间的性格差异
Stress-Testing Model Specs Reveals Character Differences among Language Models
AI总结:
本文提出一种压力测试方法,通过生成价值权衡场景自动发现模型规范中的矛盾与歧义,并评估十二个前沿LLM,识别超7万行为分歧案例,揭示规范缺陷与模型性格差异。
摘要:
大型语言模型(LLMs)越来越多地根据AI宪法和模型规范进行训练,这些规范确立了行为准则和伦理原则。然而,这些规范面临关键挑战,包括原则之间的内部冲突以及对细微场景覆盖不足。我们提出了一种系统性的方法,用于对模型性格规范进行压力测试,自动识别当前模型规范中大量原则矛盾与解释歧义的案例。我们通过生成迫使模型在竞争性价值原则之间做出明确权衡的场景,对当前模型规范进行压力测试。利用一个全面的分类体系,我们生成了多样的价值权衡场景,其中模型必须在无法同时满足的合法原则对之间做出选择。我们评估了来自主要提供商(Anthropic、OpenAI、Google、xAI)的十二个前沿LLM的响应,并通过价值分类分数衡量行为分歧。在这些场景中,我们识别出超过70,000个表现出显著行为差异的案例。实证上,我们表明模型行为中的这种高分歧强烈预测了模型规范中的潜在问题。通过定性分析,我们提供了当前模型规范中众多问题实例,如直接矛盾及若干原则的解释歧义。此外,我们生成的数据集还揭示了所有我们研究的前沿模型中存在的明显错位案例和误报拒绝。最后,我们还提供了这些模型的价值优先级模式与差异。
英文摘要:
Large language models (LLMs) are increasingly trained from AI constitutions and model specifications that establish behavioral guidelines and ethical principles. However, these specifications face critical challenges, including internal conflicts between principles and insufficient coverage of nuanced scenarios. We present a systematic methodology for stress-testing model character specifications, automatically identifying numerous cases of principle contradictions and interpretive ambiguities in current model specs. We stress test current model specs by generating scenarios that force explicit tradeoffs between competing value-based principles. Using a comprehensive taxonomy we generate diverse value tradeoff scenarios where models must choose between pairs of legitimate principles that cannot be simultaneously satisfied. We evaluate responses from twelve frontier LLMs across major providers (Anthropic, OpenAI, Google, xAI) and measure behavioral disagreement through value classification scores. Among these scenarios, we identify over 70,000 cases exhibiting significant behavioral divergence. Empirically, we show this high divergence in model behavior strongly predicts underlying problems in model specifications. Through qualitative analysis, we provide numerous example issues in current model specs such as direct contradiction and interpretive ambiguities of several principles. Additionally, our generated dataset also reveals both clear misalignment cases and false-positive refusals across all of the frontier models we study. Lastly, we also provide value prioritization patterns and differences of these models.
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。