提示词应用实践

学习目标

本课程通过LLMs生成情感分析样本和标注葡萄9品鉴数据，展示了其文本生成和数据标注能力。同时，利用PAL模型解决日期计算问题，学习了LLMs与编程运行时结合实现复杂推理的方法，为自然语言处理应用提供了实践经验。

相关知识点

提示词应用实践

学习内容

1 提示词应用实践

提示词的定义

提示词是给AI的指令，帮助它理解你的需求。比如，如果你想让AI写一个故事，你可以给它一个提示词：“写一个关于勇敢骑士的故事”。好的提示词能让AI更快地给出你想要的结果。提示词越具体，AI生成的内容就越符合你的期望。

提示词的使用

使用提示词时，要尽量简洁明了。 例如，如果你想让AI翻译一段文字，可以直接说：“把这段英文翻译成中文”。同时，提供足够的上下文也很重要。比如，如果你想让AI写一个科幻故事，可以加上背景信息：“在一个未来的世界里，人类已经可以自由穿梭于星际之间”。这样，AI就能更好地理解你的需求。

提示词的优化

如果第一次的结果不满意，可以尝试调整提示词。 比如，如果你觉得AI生成的故事不够有趣，可以换一个说法：“写一个关于勇敢骑士的冒险故事，情节要充满惊喜”。多试几次，找到最适合你需求的表达方式。此外，多观察AI的输出，根据结果调整提示词，这样可以逐步优化结果。

1.1 LLMs生成文本及标注案例

LLMs具有很强的生成文本的能力。使用有效的提示策略可以引导模型产生更好、更一致和更真实的响应。LLMs 对于生成数据也特别有用，这对于运行各种实验非常有用。例如，我们可以使用它来为情感分类器生成快速样本，如下所示：

提示:

生成10个情感分析的示例。示例分为正面和负面两类。生成2个负面示例和8个正面示例。使用以下格式生成示例:Q: <句子>
A: <情绪>

输出:

Q: 我刚刚得到了最好的消息！
A: 积极Q: 我们刚刚得到了加薪！
A: 积极Q: 我为我今天所取得的成就感到自豪。
A: 积极Q: 今天是我有史以来最棒的一天！
A: 积极Q: 我真的很期待周末的到来。
A: 积极Q: 我刚刚收到了最好的礼物！
A: 积极Q: 我现在好开心。
A: 积极Q: 我非常幸运有一个这么棒的家人。
A: 积极Q: 外面的天气真阴沉。
A: 消极Q: 我刚刚得到了一个可怕的消息。
A: 消极

这是一个关于葡萄9品鉴描述的标注示例，它通过标注葡萄9的口感（taste）、风味（flavor）和香气（aroma）相关词汇及其在文本中的位置，用于自然语言处理（NLP）中的命名实体识别（NER）任务。

提示:

produce 3 wine reviews and label taste, flavor, aroma related token; present the result as a json file, in addition add the coordinate of each term for NER task

输出:

[{"review": "This Chardonnay has a rich, creamy texture with notes of vanilla and buttery oak. The taste is smooth and well-balanced, with a hint of tropical fruit on the finish.","tokens": [{"text": "rich", "label": "flavor", "start": 5, "end": 9},{"text": "creamy", "label": "texture", "start": 10, "end": 16},{"text": "vanilla", "label": "flavor", "start": 28, "end": 35},{"text": "buttery", "label": "flavor", "start": 40, "end": 47},{"text": "oak", "label": "aroma", "start": 48, "end": 51},{"text": "smooth", "label": "taste", "start": 61, "end": 67},{"text": "well-balanced", "label": "taste", "start": 68, "end": 81},{"text": "hint", "label": "flavor", "start": 89, "end": 93},{"text": "tropical fruit", "label": "flavor", "start": 98, "end": 112},{"text": "finish", "label": "taste", "start": 116, "end": 122}]},{"review": "This Cabernet Sauvignon has a bold, full-bodied flavor with notes of black cherry and dark chocolate. The aroma is rich and complex, with hints of tobacco and leather.","tokens": [{"text": "bold", "label": "flavor", "start": 5, "end": 9},{"text": "full-bodied", "label": "texture", "start": 10, "end": 21},{"text": "black cherry", "label": "flavor", "start": 30, "end": 42},{"text": "dark chocolate", "label": "flavor", "start": 47, "end": 60},{"text": "rich", "label": "aroma", "start": 69, "end": 73},{"text": "complex", "label": "aroma", "start": 78, "end": 85},{"text": "hints", "label": "aroma", "start": 90, "end": 96},{"text": "tobacco", "label": "aroma", "start": 101, "end": 108},{"text": "leather", "label": "aroma", "start": 113, "end": 120}]},{"review": "This Riesling has a crisp, refreshing taste with notes of green apple and honey. The aroma is floral and fruity, with a hint of citrus.","tokens": [{"text": "crisp", "label": "texture", "start": 5, "end": 10},{"text": "refreshing", "label": "texture", "start": 12, "end": 22},{"text": "green apple", "label": "flavor", "start": 31, "end": 42},]}

1.2 PAL模型日期计算示例

如下展示了一种创新方法，利用大型语言模型（LLMs）来读取自然语言问题，并生成程序作为中间推理步骤。这种被称为程序辅助语言模型（PAL）的方法与传统的思维链提示有所不同。PAL不是通过自由格式的文本来获取解决方案，而是将解决方案的执行步骤委托给编程运行时环境，例如Python解释器。
这个应用程序处理的是一个日期计算问题，它告诉大模型当前的日期以及用户的年龄，然后让大模型推断出用户的出生日期。

1.2.1 安装依赖

%pip install httpx
%pip install openai

1.2.2 初始化大模型相关参数

import os
import httpx
from openai import OpenAI# 此课程基于DeepSeek-R1运行，学员可自行前往官网进行api申请
client = OpenAI(base_url=<BASE_URL>,api_key=<API_KEY>,http_client=httpx.Client(verify=False),
)

1.2.3 格式化处理大模型输出的内容

from openai import ChatCompletion
def extract_after_think(chat_completion: ChatCompletion) -> str:"""从 ChatCompletion 对象中提取 </think> 标签之后的内容。:param chat_completion: openai.types.chat.chat_completion.ChatCompletion 对象:return: </think> 标签之后的内容"""# 获取 ChatCompletion 对象中的 content 属性content = chat_completion.choices[0].message.content# 查找 </think> 标签的位置start_index = content.find("</think>")if start_index != -1:# 返回 </think> 标签之后的内容return content[start_index + len("</think>"):]else:# 如果没有找到 </think> 标签，返回空字符串return ""

1.2.4 调用大模型

from datetime import datetime
from dateutil.relativedelta import relativedelta#请根据实际情况替换<model_name>
def get_completion(prompt, temperature=0.6):messages = [{"role": "user", "content": prompt}]response = client.chat.completions.create(model=<model_name>,messages=messages,stream=False,temperature=temperature,max_tokens=2048,top_p=0.95)return extract_after_think(response)question = "Today is 27 February 2023. I was born exactly 25 years ago. What is the date I was born in MM/DD/YYYY?"prompt = f"""
# Q: 2015 is coming in 36 hours. What is the date one week from today in MM/DD/YYYY?
# If 2015 is coming in 36 hours, then today is 36 hours before.
today = datetime(2015, 1, 1) - relativedelta(hours=36)
# One week from today,
one_week_from_today = today + relativedelta(weeks=1)
# The answer formatted with %m/%d/%Y is
one_week_from_today.strftime('%m/%d/%Y')
# Q: The first day of 2019 is a Tuesday, and today is the first Monday of 2019. What is the date today in MM/DD/YYYY?
# If the first day of 2019 is a Tuesday, and today is the first Monday of 2019, then today is 6 days later.
today = datetime(2019, 1, 1) + relativedelta(days=6)
# The answer formatted with %m/%d/%Y is
today.strftime('%m/%d/%Y')
# Q: The concert was scheduled to be on 06/01/1943, but was delayed by one day to today. What is the date 10 days ago in MM/DD/YYYY?
# If the concert was scheduled to be on 06/01/1943, but was delayed by one day to today, then today is one day later.
today = datetime(1943, 6, 1) + relativedelta(days=1)
# 10 days ago,
ten_days_ago = today - relativedelta(days=10)
# The answer formatted with %m/%d/%Y is
ten_days_ago.strftime('%m/%d/%Y')
# Q: It is 4/19/1969 today. What is the date 24 hours later in MM/DD/YYYY?
# It is 4/19/1969 today.
today = datetime(1969, 4, 19)
# 24 hours later,
later = today + relativedelta(hours=24)
# The answer formatted with %m/%d/%Y is
today.strftime('%m/%d/%Y')
# Q: Jane thought today is 3/11/2002, but today is in fact Mar 12, which is 1 day later. What is the date 24 hours later in MM/DD/YYYY?
# If Jane thought today is 3/11/2002, but today is in fact Mar 12, then today is 3/1/2002.
today = datetime(2002, 3, 12)
# 24 hours later,
later = today + relativedelta(hours=24)
# The answer formatted with %m/%d/%Y is
later.strftime('%m/%d/%Y')
# Q: Jane was born on the last day of Feburary in 2001. Today is her 16-year-old birthday. What is the date yesterday in MM/DD/YYYY?
# If Jane was born on the last day of Feburary in 2001 and today is her 16-year-old birthday, then today is 16 years later.
today = datetime(2001, 2, 28) + relativedelta(years=16)
# Yesterday,
yesterday = today - relativedelta(days=1)
# The answer formatted with %m/%d/%Y is
yesterday.strftime('%m/%d/%Y')
# Q: {question}
""".strip()result = get_completion(prompt)
print(result)

---

# 大语言模型提示工程与应用
1. 提示工程入门指南
2. 提示词基础使用方式
3. 大语言模型进阶提示工程技术
4. LLMs文本生成与数据标注实践：情感分析与葡萄酒品鉴
5. ChatGPT提示工程技术指南
6. 大语言模型对抗性提示安全防御指南
7. 提示工程：提升模型准确性与减少偏见的方法
8. 前沿提示工程技术探索