Language models (LMs) often struggle to pay enough attention to the input
context, and generate texts that are unfaithful or contain hallucinations. To
mitigate this issue, we present context-aware decoding (CAD), which follows a
contrastive output distribution that amplifies the difference between the
output probabilities when a model is used with and without context. Our
experiments show that CAD, without additional training, significantly improves
the faithfulness of different LM families, including OPT, GPT, LLaMA and
FLAN-T5 for summarization tasks (e.g., 14.3% gain for LLaMA in factuality
metrics). Furthermore, CAD is particularly effective in overriding a model's
prior knowledge when it contradicts the provided context, leading to
substantial improvements in tasks where resolving the knowledge conflict is
essential.

提出了一种上下文感知解码技术（CAD），通过对比上下文的存在与否来放大模型生成输出的概率得分，从而显著提高了 OPT、GPT、LLaMA 和 FLAN-T5 等不同 LM 系列在摘要任务中的可信度，并能够有效地改变模型的先验知识以解决知识冲突而带来的质量问题。