Zero-Shot Prompting

零樣本提示

LLMs today trained on large amounts of data and tuned to follow instructions, are capable of performing tasks zero-shot. We tried a few zero-shot examples in the previous section. Here is one of the examples we used:

今天的LLMs在大量的資料訓練和調整後，能夠執行zero-shot任務。我們在前一節中嘗試了一些zero-shot示範。這是我們使用的其中一個示範：

Prompt:

提示:

Classify the text into neutral, negative or positive.

Text: I think the vacation is okay.
Sentiment:

Output:

輸出:

Neutral

Note that in the prompt above we didn't provide the model with any examples -- that's the zero-shot capabilities at work.

請注意，上面的提示中我們沒有向模型提供任何示範——這就是zero-shot能力在發揮作用。

Instruction tuning has shown to improve zero-shot learning Wei et al. (2022) (opens in a new tab). Instruction tuning is essentially the concept of finetuning models on datasets described via instructions. Furthermore, RLHF (opens in a new tab) (reinforcement learning from human feedback) has been adopted to scale instruction tuning wherein the model is aligned to better fit human preferences. This recent development powers models like ChatGPT. We will discuss all these approaches and methods in upcoming sections.

指令微調已被證明可以改善零樣本學習Wei et al. (2022) (opens in a new tab)。指令微調本質上是在描述指令的資料集上微調模型的概念。此外，RLHF (opens in a new tab)（從人類反饋中進行強化學習）已被採用來擴充套件指令微調，使模型更好地適應人類偏好。這一最新發展推動了像ChatGPT這樣的模型。我們將在接下來的章節中討論所有這些方法和方法。

When zero-shot doesn't work, it's recommended to provide demonstrations or examples in the prompt which leads to few-shot prompting. In the next section, we demonstrate few-shot prompting.

當zero-shot 無法運作時，建議在提示中提供示範或範例，以進行few-shot 提示。在下一節中，我們將展示few-shot 提示。