Technology Radar
Published : Sep 27, 2023
NOT ON THE CURRENT EDITION
This blip is not on the current edition of the Radar. If it was on one of the last few editions, it is likely that it is still relevant. If the blip is older, it might no longer be relevant and our assessment might be different today. Unfortunately, we simply don't have the bandwidth to continuously review blips from previous editions of the Radar.
Understand more
Sep 2023
Trial
在使用大语言模型的各种应用中,我们在 半结构化自然语言输入 方面取得了成功。结构化输入,如 JSON 文档,清晰而精确,为模型提供了所寻求响应类型的指示。以这种方式限制响应有助于缩小问题空间,并且可以产生更准确的响应,特别是当结构符合领域特定语言(DSL)的语法或模式情况下。我们还发现,将结构化输入与自然语言注释或标记结合使用,比仅使用自然语言或结构化输入产生更好的响应。通常,在构建提示时,自然语言会与结构化内容简单地交织在一起。与许多大语言模型行为一样,我们不知道为什么这样做有效,但我们的经验表明,在人工编写的代码中加入自然语言注释也会改善基于大语言模型的编码助手的输出质量。