论文标题
在语言模型中，迅速的压缩和对比度调节，以减少可控性和毒性
Prompt Compression and Contrastive Conditioning for Controllability and Toxicity Reduction in Language Models
论文作者
论文摘要
我们探讨了压缩用于调节语言模型的提示的想法，并表明压缩提示可以保留有关原始提示的大量信息。对于严重压缩的提示，虽然丢失了细粒度的信息，但抽象信息和一般情感可以以惊人的参数保留，这在解释时间算法的上下文中可用于可控性和降低毒性。我们探讨了将语言模型生成转向理想文本的对比条件，并远离不良文本，并发现某些复杂的提示可以被有效地压缩到单个令牌中以指导生成。我们还表明，压缩提示在很大程度上是组成的，并且可以构造，以便可以用来控制生成的文本的独立方面。
We explore the idea of compressing the prompts used to condition language models, and show that compressed prompts can retain a substantive amount of information about the original prompt. For severely compressed prompts, while fine-grained information is lost, abstract information and general sentiments can be retained with surprisingly few parameters, which can be useful in the context of decode-time algorithms for controllability and toxicity reduction. We explore contrastive conditioning to steer language model generation towards desirable text and away from undesirable text, and find that some complex prompts can be effectively compressed into a single token to guide generation. We also show that compressed prompts are largely compositional, and can be constructed such that they can be used to control independent aspects of generated text.
