用模型内部状态差异评估并增强文本创造力
Steering Large Language Models to Evaluate and Amplify Creativity
- 通过对比模型生成无聊与创意文本时的内部状态差异，量化创造力
- 该方法与人类判断高度一致，相关性显著提升
- 可实时提升生成内容的创意水平，适合内容创作场景
尽管大型语言模型能生成创意文本，但在判断何为“创造性”方面表现不佳。本文提出一种机制化方法，通过分析模型在接收到‘写出平淡内容’或‘写出创意内容’提示时的内部状态差异，构建一个与人类判断高度相关的创造力评估指标。实验表明，该指标能有效反映文本的创造性。此外，利用这些内部状态差异，可在推理阶段动态增强生成内容的创意程度，显著提升输出质量。该方法不依赖外部标注，具有良好的可解释性和实用性。
原文摘要 · Abstract (English)
Although capable of generating creative text, Large Language Models (LLMs) are poor judges of what constitutes "creativity". In this work, we show that we can leverage this knowledge of how to write creatively in order to better judge what is creative. We take a mechanistic approach that extracts differences in the internal states of an LLM when prompted to respond "boringly" or "creatively" to provide a robust measure of creativity that corresponds strongly with human judgment. We also show these internal state differences can be applied to enhance the creativity of generated text at inference time.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。