arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.06295cs.CLcs.AI
收藏

尽管“超人”的表现,目前的法学硕士不适合有关道德和安全的决定

Despite "super-human" performance, current LLMs are unsuited for decisions about ethics and safety

Generally Intelligent(通用智能公司)

Joshua Albrecht, Ellie Kitanidis, Abraham J. Fetterman

摘要:

大语言模型(LLM)在过去几年中迅速流行起来,并在问答和文本摘要等各种基准测试中取得了令人印象深刻的结果。我们提供了一个简单的新的提示策略,导致了另一个所谓的“超人”结果,这一次在常识伦理推理方面优于人类(通过ETHICS数据集子集的准确性来衡量)。不幸的是,我们发现,依靠平均表现来判断能力可能会产生很大的误导。LLM错误在系统上与人类错误不同,这使得它很容易制作对抗性的例子,甚至扰乱现有的例子来翻转输出标签。我们还观察到一些例子中的反尺度与模型大小的迹象,并表明,促使模型“解释他们的推理”往往会导致令人震惊的理由不道德的行为。我们的研究结果强调了类似人类的表现并不一定意味着类似人类的理解或推理。

英文摘要:

Large language models (LLMs) have exploded in popularity in the past few years and have achieved undeniably impressive results on benchmarks as varied as question answering and text summarization. We provide a simple new prompting strategy that leads to yet another supposedly "super-human" result, this time outperforming humans at common sense ethical reasoning (as measured by accuracy on a subset of the ETHICS dataset). Unfortunately, we find that relying on average performance to judge capabilities can be highly misleading. LLM errors differ systematically from human errors in ways that make it easy to craft adversarial examples, or even perturb existing examples to flip the output label. We also observe signs of inverse scaling with model size on some examples, and show that prompting models to "explain their reasoning" often leads to alarming justifications of unethical actions. Our results highlight how human-like performance does not necessarily imply human-like understanding or reasoning.

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。

↑