Contrastive Perplexity for Controlled Generation: An Application in Detoxifying Large Language Models
CoRR(2024)
摘要
The generation of undesirable and factually incorrect content of large
language models poses a significant challenge and remains largely an unsolved
issue. This paper studies the integration of a contrastive learning objective
for fine-tuning LLMs for implicit knowledge editing and controlled text
generation. Optimizing the training objective entails aligning text
perplexities in a contrastive fashion. To facilitate training the model in a
self-supervised fashion, we leverage an off-the-shelf LLM for training data
generation. We showcase applicability in the domain of detoxification. Herein,
the proposed approach leads to a significant decrease in the generation of
toxic content while preserving general utility for downstream tasks such as
commonsense reasoning and reading comprehension. The proposed approach is
conceptually simple but empirically powerful.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要