We call into question the recently popularized method of direct model editing as a means of correcting factual errors in LLM generations. We contrast model editing with three similar but distinct approaches that pursue better defined objectives: (1) retrieval-based architectures, which decouple factual memory from inference and linguistic capabilities embodied in LLMs; (2) concept erasure methods, which aim at preventing systemic bias in generated text; and (3) attribution methods, which aim at grounding generations into identified textual sources. We argue that direct model editing cannot be trusted as a systematic remedy for the disadvantages inherent to LLMs, and while it has proven potential in improving model explainability, it opens risks by reinforcing the notion that models can be trusted for factuality. We call for cautious promotion and application of model editing as part of the LLM deployment process, and for responsibly limiting the use cases of LLMs to those not relying on editing as a critical component.

翻译：我们质疑近期流行的直接模型编辑方法——即将其作为纠正大语言模型（LLM）生成内容中事实错误的手段。本文将模型编辑与三种目标更明确的相似但不同的方法进行对比：（1）基于检索的架构，将事实记忆与大语言模型中蕴含的推理及语言能力解耦；（2）概念擦除方法，旨在防止生成文本中的系统性偏见；（3）归因方法，旨在将生成内容锚定于可识别的文本来源。我们认为，直接模型编辑不能作为系统性地弥补大语言模型固有缺陷的可靠方案；尽管它在提升模型可解释性方面已展现出潜力，但强化了"模型在事实性上值得信赖"的观念，由此带来风险。我们呼吁在部署大语言模型时审慎推广和应用模型编辑，并负责任地将大语言模型的使用场景限制在不依赖编辑作为关键组件的范围内。