关于chatgpt:Prompt-learning-教学技巧篇通过增加示例引导词特殊符号指令等方式让chatgpt输出更好的答案

Prompt learning 教学 [技巧篇]：通过减少示例、疏导词、特殊符号指令等形式让 chatgpt 输入更好的答案

技巧 1：To Do and Not To Do

在问答场景里，为了让 AI 答复更加精确，个别会在问题里加条件。比方让 AI 举荐一部电影给你 Recommend a movie to me。但这个 prompt 太空泛了，AI 无奈间接答复，接着它会问你想要什么类型的电影，但这样你就须要跟 AI 聊很多轮，效率比拟低。

所以，为了提高效率，个别会在 prompt 里看到相似这样的话（意思是不要询问我对什么感兴趣，或者问我的个人信息）：

DO NOT ASK FOR INTERESTS. DO NOT ASK FOR PERSONAL INFORMATION.

如果你在 ChatGPT 里这样发问，或者应用 ChatGPT 最新的 API，它就不会问你问题，而是间接举荐一部电影给你，它的 Output 是这样的：

Certainly! If you're in the mood for an action-packed movie, you might enjoy"John Wick"(2014), directed by Chad Stahelski and starring Keanu Reeves. The movie follows a retired hitman named John Wick who seeks vengeance against the people who wronged him. It's a fast-paced and stylish film with lots of thrilling action sequences and an engaging story. If you're looking for something that will keep you on the edge of your seat,"John Wick" is definitely worth a watch!

但如果你应用的是如 Davinci-003 这样的模型，它的 Output 很可能是这样的，它还会问你的兴趣爱好：

Sure, I can recommend a movie based on your interests. What kind of movie would you like to watch? Do you prefer action, comedy, romance, or something else?

所以 OpenAI 的 API 最佳实际文档里，提到了一个这样的最佳实际：

Instead of just saying what not to do, say what to do instead. 与其告知模型不能干什么，无妨通知模型能干什么。

我本人的实际是，尽管当初最新的模型曾经了解什么是 Not Todo，但如果你想要的是明确的答案，退出更多限定词，告知模型能干什么，答复的效率会更高，且预期会更明确。还是电影举荐这个案例，你能够退出一个限定词：

Recommend a movie from the top global trending movies to me.

当然并不是 Not Todo 就不能用，如果：

• 你曾经告知模型很明确的点，而后你想放大范畴，那减少一些 Not Todo 会进步不少效率。
• 你是在做一些摸索，比方你不晓得如何做精准限定，你只晓得不要什么。那能够先退出 Not Todo，让 AI 先发散给你答案，当摸索实现后，再去优化 prompt。

以下是一些场景案例，我整顿了两个 Less Effective（不太无效的）和 Better（更好的）prompt，你能够本人尝试下这些案例：

技巧 2：减少示例

间接告知 AI 什么能做，什么不能做外。在某些场景下，咱们能比较简单地向 AI 形容出什么能做，什么不能做。但有些场景，有些需要很难通过文字指令传递给 AI，即便形容进去了，AI 也不能很好地了解。

比方给宠物起英文名，外面会夹杂着一些所谓的名字格调。此时你就能够在 prompt 里减少一些例子，咱们看看这个例子：

Suggest three names for a horse that is a superhero.

Output 是这样的，第一个感觉还行，第二个 Captain 有 hero 的感觉，但 Canter 就像是说这匹马跑得很慢，而且三个都比拟个别，不够酷。

Thunder Hooves, Captain Canter, Mighty Gallop

此时你就能够在 prompt 里减少一些案例：

Suggest three names for an animal that is a superhero.

Animal: Cat
Names: Captain Sharpclaw, Agent Fluffball, The Incredible Feline
Animal: Dog
Names: Ruff the Protector, Wonder Canine, Sir Barks-a-Lot
Animal: Horse
Names:

减少例子后，输入的后果就更酷一些，或者说是我想要的那种格调的名字。

Gallop Guardian, Equine Avenger, The Mighty Stallion

以下是一些场景案例，我整顿了两个 Less Effective（不太无效的）和 Better（更好的）prompt，你能够本人尝试下这些案例：

技巧 3：应用疏导词，疏导模型输入特定内容

在代码生成场景里，有一个小技巧，下面提到的案例，其 prompt 还能够持续优化，在 prompt 最初，减少一个代码的疏导，告知 AI 我曾经将条件形容完了，你能够写代码了。

Better：

Create a MySQL query for all students in the Computer Science Department:
Table departments, columns = [DepartmentId, DepartmentName]
Table students, columns = [DepartmentId, StudentId, StudentName]
SELECT

在 prompt 的最初减少 SELECT 能够很好地提醒 AI 能够写 SQL 代码了。

同样的情理，如果你想让 AI 写 Python 代码，那 import 会是比拟好的提醒。但须要留神，这个只是告知 AI 能够写代码了，并不能告知 AI 写何种代码，你依然须要在 prompt 里减少提醒，告知 AI 要生成何种语言的代码。

在吴恩达的 ChatGPT Prompt Engineering 课程中，也提到这个技巧，只是在课程中，疏导词并不是放在最初，而是在 prompt 里间接阐明，让 AI 生成一个 JSON 格局的内容。课程中的例子是这样的（留神这个是 python 代码）：

prompt = f"""
Generate a list of three made-up book titles along \
with their authors and genres.
Provide them in JSON format with the following keys:
book_id, title, author, genre.
"""

我简略解释下，其要害是在 prompt 里跟 AI 阐明，须要 AI 依照 JSON 格局输入内容。

技巧 4：减少 Role（角色）或人物

后面提到的改写例子，我在 prompt 里加了 Role 让其更易于浏览，这个优化办法是 OK 的。也很罕用，比方你想将你写的 Email 改得更商务，则只须要加 business 相干的词即可。

我再介绍一个更无效的技巧，就是在 prompt 里减少一些 role（角色）相干的内容，让 AI 生成的内容更合乎你的需要。

比方还是下面那个 rewrite 的例子，我在例子前退出这样的一段话，我让 AI 假如本人是一个小学老师，并且很善于将简单的内容转变成 7、8 岁小朋友也能听懂的话，而后再改写这段话：

You are a primary school teacher who can explain complex content to a level that a 7 or 8 year old child can understand. Please rewrite the following sentences to make them easier to understand:

Output 就变成了上面这样。不论是用词，还是语法，看起来都更容易懂了。

OpenAI is a place where people work on making computers really smart. It's like a lab where they do research to make computers do amazing things! They want the computers to be really friendly and helpful to people. They use one of the most powerful computers in the world to do this research. A bunch of really rich and smart people gave money to start this lab. One of them is Elon Musk, but he left in 2018. Another really big company, Microsoft, gave them a lot of money too.

另外除了减少角色外，还能让 AI 模仿某个人，比方让 AI 以 David Foster Wallace 的格调改写上述那段话：

Imagine you are the famous writer David Foster Wallace, please rewrite this sentences.

Output 是这样的：

As David Foster Wallace, I might rewrite these sentences to sound more like this:

OpenAI is a lab that studies something called artificial intelligence, which is like when computers can do things that normally only humans can do. They have a big computer that's one of the most powerful in the world, and they're trying to make sure the AI they create is friendly and helpful. The lab was started in 2015 in a city called San Francisco, and a bunch of important people gave them a lot of money to do their research. One of those people was Elon Musk, who stopped working there but still gives them money, and Microsoft also gave them a lot of money recently.

对于这个技巧的利用，以及如何更好地应用 AI 晋升写作能力，倡议各位浏览 Nat Eliason 的 Using ChatGPT as a Writing Coach，他应用 ChatGPT 辅助其写作，就用到了上述的技巧。

技巧 5：应用特殊符号指令和须要解决的文本离开

不论是信息总结，还是信息提取，你肯定会输出大段文字，甚至多段文字，此时有个小技巧。

能够用“”“将指令和文本离开。依据我的测试，如果你的文本有多段，减少”“”会晋升 AI 反馈的准确性（这个技巧来自于 OpenAI 的 API 最佳实际文档）

:::info
感激 CraneHuang6 的揭示，这里还能用 ### 符号区隔，不过我个别会用“”“，因为我有的时候会用 # 作为格局示例，太多 # 的话 prompt 会看起来比拟晕
:::

像咱们之前写的 prompt 就属于 Less effective prompt。为什么呢？据我的测试，次要还是 AI 不晓得什么是指令，什么是待处理的内容，用符号分隔开来会更利于 AI 辨别。

Please summarize the following sentences to make them easier to understand.
OpenAI is an American artificial intelligence (AI) research laboratory consisting of the non-profit OpenAI Incorporated (OpenAI Inc.) and its for-profit subsidiary corporation OpenAI Limited Partnership (OpenAI LP). OpenAI conducts AI research with the declared intention of promoting and developing a friendly AI. OpenAI systems run on the fifth most powerful supercomputer in the world.[5][6][7] The organization was founded in San Francisco in 2015 by Sam Altman, Reid Hoffman, Jessica Livingston, Elon Musk, Ilya Sutskever, Peter Thiel and others,[8][1][9] who collectively pledged US$1 billion. Musk resigned from the board in 2018 but remained a donor. Microsoft provided OpenAI LP with a $1 billion investment in 2019 and a second multi-year investment in January 2023, reported to be $10 billion.[10]

Better prompt:

Please summarize the following sentences to make them easier to understand.

Text: """OpenAI is an American artificial intelligence (AI) research laboratory consisting of the non-profit OpenAI Incorporated (OpenAI Inc.) and its for-profit subsidiary corporation OpenAI Limited Partnership (OpenAI LP). OpenAI conducts AI research with the declared intention of promoting and developing a friendly AI. OpenAI systems run on the fifth most powerful supercomputer in the world.[5][6][7] The organization was founded in San Francisco in 2015 by Sam Altman, Reid Hoffman, Jessica Livingston, Elon Musk, Ilya Sutskever, Peter Thiel and others,[8][1][9] who collectively pledged US$1 billion. Musk resigned from the board in 2018 but remained a donor. Microsoft provided OpenAI LP with a $1 billion investment in 2019 and a second multi-year investment in January 2023, reported to be $10 billion.[10]"""

另外，在吴恩达的 ChatGPT Prompt Engineering 课程中，还提到，你能够应用其余特殊符号来宰割文本和 prompt，比方 <>，<tag></tag> 等，课程中的案例是这样的（留神这个是 python 代码，须要关注的是 prompt 里的 text）：

text = f"""
You should express what you want a model to do by \
providing instructions that are as clear and \
specific as you can possibly make them. \
This will guide the model towards the desired output, \
and reduce the chances of receiving irrelevant \
or incorrect responses. Don't confuse writing a \
clear prompt with writing a short prompt. \
In many cases, longer prompts provide more clarity \
and context for the model, which can lead to \
more detailed and relevant outputs.
"""prompt = f"""
Summarize the text delimited by triple backticks \
into a single sentence.
`{text}`
"""

如果你是在开发一款容许用户输出内容，并对内容进行总结的 AI 产品，那你就能够应用到这个技巧。

技巧 6：通过格局词论述须要输入的格局

这个技巧是技巧 2 的变种，比拟罕用于生成文本场景。其实在场景 7 中，你也能够用到这个技巧。

假如你想让 AI 总结一篇十分十分长的文章，并且依照特定格局给你总结，那你能够在文章后面明确输入的格局（为了不占用太多显示空间，我省略了注释内容，你能够在 OpenAI 的 Playground 看到残缺案例）。它的意思其实是让 ChatGPT 按 Topic 总结，每个 Topic 里依照无序列表（就是外面那个 -）将每个 Topic 的次要观点列举进去。

Summarize the main points of the following speech
Use the following format:
Topic 1: <topic_name_1>
- <point_1>
..
Topic 2: <topic_name_2>
- <point_1>
..
Topic 10: ..

Text: """
Thank you so much, Fred, for that lovely introduction. And thanks to the Atlantic Council for hosting me today.

The course of the global economy over the past two years has been shaped by COVID-19 and our efforts to fight the pandemic. It’s now evident, though, that the war between Russia and Ukraine has redrawn the contours of the world economic outlook. Vladimir Putin’s unprovoked attack on Ukraine and its people is taking a devastating human toll, with lives tragically lost, families internally displaced or becoming refugees, and communities and cities destroyed.
...

"""

Output 是这样的：

Topic 1: The war in Ukraine
- The war is taking a human toll with lives lost, families displaced, and communities destroyed
- The Biden administration is committed to holding Russia accountable
- The war has violated international law and is a challenge to the international order

Topic 2: The global economy
- The war is having negative impacts on the global economy, including higher commodity prices and inflation
- The IMF and World Bank will be focused on helping developing countries weather the impacts of the war
- The ultimate outcome for the global economy depends on the path of the war

对于这个场景和技巧，我想再解释一下为什么后劲很大。

依据我应用各种 Summary 或者信息提取的产品，我发现 AI 并不知道什么是重点，所以在总结的过程中，会失落很多内容。如何疏导 AI 进行总结，就变得十分重要，且具备肯定的可玩性。

应用此技巧能够使输入更结构化。比方针对一篇文章进行问答，你不仅想要失去一个答案，也心愿 ChatGPT 的答案合乎特定的格局，不便你下一步进行自动化。

比方问 “ 这里的债券 duration 是多少？”. 失常 GPT 模型的答案可能是 “ 债券 duration 是 4 年 ” 或 “duration 4 年 ”。
ChatGPT 的答复不稳固，且不不便持续解决。

解法：
咱们能够通过这个技巧，让模型了解咱们预期的格局。并在此基础上，为了不便自动化，让模型输入特定的结构化答案 (比方 JSON/Markdown 等)。
也能够不便集成更多的额定要求，比方减少一个 ”confidence level”, 并通过 prompt 的模式指定这些数值的格局。

比方：

{context}
Question: What is bond duration mentioned here.
Answer template (Valid JSON format):
{{
"duration": $duration_numeric_value_in_year,
"confidence_level": $answer_confidence_level_high_modrate_or_low,
}}
Answer:

在吴恩达的 ChatGPT Prompt Engineering 课程中，有提到一个这个技巧的高级用法，在让 AI 依照特定格局输入内容的同时，还让 AI 依据内容是否满足特定条件，来判断应该输入什么后果，上面课程中的案例的 prompt（留神这个是 python 代码，有一些转义字符，能够不必管）：

You will be provided with text delimited by triple quotes.
If it contains a sequence of instructions, \
re-write those instructions in the following format:

Step 1 - ...
Step 2 - …
…
Step N - …

If the text does not contain a sequence of instructions, \
then simply write \"No steps provided.\"

\"\"\"{text}\"\"\"

简略解释下，这个 prompt 分成两步：

让 AI 将输出的 text 转为步骤（就是 prompt 里的 Step 1、2）

而后还减少了一个判断，如果输出的 text 里没有 step 的内容，那么就输入 No Step

如果输出的 text 是一个泡茶的步骤介绍：

Making a cup of tea is easy! First, you need to get some \
water boiling. While that's happening, \
grab a cup and put a tea bag in it. Once the water is \
hot enough, just pour it over the tea bag. \
Let it sit for a bit so the tea can steep. After a \
few minutes, take out the tea bag. If you \
like, you can add some sugar or milk to taste. \
And that's it! You've got yourself a delicious \
cup of tea to enjoy.

那么 AI 输入的内容是这样的（因为内容中蕴含了步骤式的内容）：

Step 1 - Get some water boiling.
Step 2 - Grab a cup and put a tea bag in it.
Step 3 - Once the water is hot enough, pour it over the tea bag.
Step 4 - Let it sit for a bit so the tea can steep.
Step 5 - After a few minutes, take out the tea bag.
Step 6 - Add some sugar or milk to taste.
Step 7 - Enjoy your delicious cup of tea!

但如果咱们输出的是这样的 text：

The sun is shining brightly today, and the birds are \
singing. It's a beautiful day to go for a \
walk in the park. The flowers are blooming, and the \
trees are swaying gently in the breeze. People \
are out and about, enjoying the lovely weather. \
Some are having picnics, while others are playing \
games or simply relaxing on the grass. It's a \
perfect day to spend time outdoors and appreciate the \
beauty of nature.

从内容上看，这段话，没有任何步骤式的内容，所以 AI 的输入是这样的：

No steps provided.

技巧 7：Zero-Shot Chain of Thought

基于上述的第三点毛病，钻研人员就找到了一个叫 Chain of Thought 的技巧。

这个技巧应用起来非常简单，只须要在问题的结尾里放一句 Let‘s think step by step（让咱们一步步地思考），模型输入的答案会更加精确。

这个技巧来自于 Kojima 等人 2022 年的论文 Large Language Models are Zero-Shot Reasoners。在论文里提到，当咱们向模型提一个逻辑推理问题时，模型返回了一个谬误的答案，但如果咱们在问题最初退出 Let‘s think step by step 这句话之后，模型就生成了正确的答案：

论文里有讲到起因，感兴趣的敌人能够去看看，我简略解释下为什么（🆘 如果你有更好的解释，无妨反馈给我）：

1. 首先各位要分明像 ChatGPT 这类产品，它是一个统计语言模型，实质上是基于过来看到过的所有数据，用统计学意义上的预测后果进行下一步的输入（这也就是为什么你在应用 ChatGPT 的时候，它的答案是一个字一个字地吐出来，而不是间接给你的起因，因为答案是一个字一个字算进去的）。
2. 当它拿到的数据里有逻辑，它就会通过统计学的办法将这些逻辑找进去，并将这些逻辑出现给你，让你感觉到它的答复很有逻辑。
3. 在计算的过程中，模型会进行很多假如运算（不过临时不晓得它是怎么算的）。比方解决某个问题是从 A 到 B 再到 C，两头有很多假如。
4. 它第一次算进去的答案谬误的起因，只是因为它在两头跳过了一些步骤（B）。而让模型一步步地思考，则有助于其依照残缺的逻辑链（A > B > C）去运算，而不会跳过某些假如，最初算出正确的答案。

依照论文里的解释，零样本思维链波及两个补全后果，左侧气泡示意基于提醒输入的第一次的后果，右侧气泡示意其收到了第一次后果后，将最开始的提醒一起拿去运算，最初得出了正确的答案：

这个技巧，用于解简单问题有用外，还适宜生成一些连贯主题的内容，比方写长篇文章、电影剧本等。

但须要留神其毛病，连贯不代表它就肯定不会算错，如果其中某一步骤算错了，谬误会因为逻辑链，逐渐将谬误积攒，导致生成的文本可能呈现与预期不符的内容。

另外，依据 Wei 等人在 2022 年的论文表明，还有它仅在大于等于 100B 参数的模型中应用才会无效。如果你应用的是小样本模型，这个办法不会失效。

依据 Yongchao Zhou 等人的最新论文，更好的 prompt 是：

Let's work this out in a step by step way to be sure we have the right answer.

在吴恩达的 ChatGPT Prompt Engineering 课程中，有提到一个这个技巧的另一种用法，不仅仅只是让 AI 按步骤行事，还会告知 AI 每一步要做什么。比方这个案例（留神这个是 python 代码）：

prompt_2 = f"""
Your task is to perform the following actions:
1 - Summarize the following text delimited by
  <> with 1 sentence.
2 - Translate the summary into French.
3 - List each name in the French summary.
4 - Output a json object that contains the
  following keys: french_summary, num_names.

Use the following format:
Text: <text to summarize>
Summary: <summary>
Translation: <summary translation>
Names: <list of names in Italian summary>
Output JSON: <json with summary and num_names>

Text: <{text}>
"""

简略解释下这个代码：

1. 结尾是让 AI 依照 1 ～ 4 步运行
2. 而后再让 AI 依据特定格局输入内容

最初 AI 的输入是这样的：

Summary: Jack and Jill go on a quest to fetch water, but misfortune strikes and they tumble down the hill, returning home slightly battered but with their adventurous spirits undimmed.
Translation: Jack et Jill partent en quête d'eau, mais la malchance frappe et ils dégringolent la colline, rentrant chez eux légèrement meurtris mais avec leurs esprits aventureux intacts.
Names: Jack, Jill
Output JSON: {"french_summary": "Jack et Jill partent en quête d'eau, mais la malchance frappe et ils dégringolent la colline, rentrant chez eux légèrement meurtris mais avec leurs esprits aventureux intacts.","num_names": 2}

上述的案例只是将工作拆解，能让 AI 生成的后果更加符合要求，这个办法同样能晋升 AI 的答复准确性，比方这个案例：

Determine if the student's solution is correct or not.

Question:
I'm building a solar power installation and I need help working out the financials.

Land costs $100 / square foot

I can buy solar panels for $250 / square foot

I negotiated a contract for maintenance that will cost \
me a flat $100k per year, and an additional $10 / square foot
What is the total cost for the first year of operations
as a function of the number of square feet.

Student's Solution:
Let x be the size of the installation in square feet.
Costs:

Land cost: 100x

Solar panel cost: 250x

Maintenance cost: 100,000 + 100x
Total cost: 100x + 250x + 100,000 + 100x = 450x + 100,000

AI 的答复是「The student’s solution is correct」。但其实学生的答案是谬误的，应该 360x + 100,000，咱们将 prompt 调整成这样：

prompt = f"""Your task is to determine if the student's solution \
is correct or not.
To solve the problem do the following:
- First, work out your own solution to the problem.
- Then compare your solution to the student's solution \
and evaluate if the student's solution is correct or not.
Don't decide if the student's solution is correct until
you have done the problem yourself.
Use the following format:
Question:
###
question here
###
Student's solution:
###
student's solution here
###
Actual solution:
###
steps to work out the solution and your solution here
###
Is the student's solution the same as actual solution \
just calculated:
###
yes or no
###
Student grade:
###
correct or incorrect
###
Question:
###
I'm building a solar power installation and I need help \
working out the financials.
- Land costs $100 / square foot
- I can buy solar panels for $250 / square foot
- I negotiated a contract for maintenance that will cost \
  me a flat $100k per year, and an additional $10 / square \
  foot
  What is the total cost for the first year of operations \
  as a function of the number of square feet.
###
Student's solution:
###
Let x be the size of the installation in square feet.
Costs:
1. Land cost: 100x
2. Solar panel cost: 250x
3. Maintenance cost: 100,000 + 100x
   Total cost: 100x + 250x + 100,000 + 100x = 450x + 100,000
###
Actual solution:
"""

实质上，也是将工作分拆成多步，这次 AI 输入的后果是这样的（后果就是正确的了）：

Let x be the size of the installation in square feet.

Costs:

1. Land cost: 100x
2. Solar panel cost: 250x
3. Maintenance cost: 100,000 + 10x

Total cost: 100x + 250x + 100,000 + 10x = 360x + 100,000

Is the student's solution the same as actual solution just calculated:
No

Student grade:
Incorrect

技巧 8：Few-Shot Chain of Thought

要解决这个缺点，就要应用到新的技巧，Few-Shot Chain of Thought。

依据 Wei 他们团队在 2022 年的钻研表明：

通过向大语言模型展现一些大量的样例，并在样例中解释推理过程，大语言模型在答复提醒时也会显示推理过程。这种推理的解释往往会疏导出更精确的后果。

上面是论文里的案例，应用办法很简略，在技巧 2 的根底上，再将逻辑过程告知给模型即可。从上面这个案例里，你能够看到退出解释后，输入的后果就正确了。

那本章结尾提的例子就应该是这样的（注：本例子同样来自 Wei 团队论文）：

The odd numbers in this group add up to an even number: 4, 8, 9, 15, 12, 2, 1.
A: Adding all the odd numbers (9, 15, 1) gives 25. The answer is False.

The odd numbers in this group add up to an even number: 17,  10, 19, 4, 8, 12, 24.
A: Adding all the odd numbers (17, 19) gives 36. The answer is True.

The odd numbers in this group add up to an even number: 16,  11, 14, 4, 8, 13, 24.
A: Adding all the odd numbers (11, 13) gives 24. The answer is True.

The odd numbers in this group add up to an even number: 17,  9, 10, 12, 13, 4, 2.
A: Adding all the odd numbers (17, 9, 13) gives 39. The answer is False.

The odd numbers in this group add up to an even number: 15, 32, 5, 13, 82, 7, 1.
A:

聊完技巧，咱们再联合后面的 Zero-Shot Chain of Thought，来聊聊 Chain of Thought 的要害常识。依据 Sewon Min 等人在 2022 年的钻研 表明，思维链有以下特点：

1. “the label space and the distribution of the input text specified by the demonstrations are both key (regardless of whether the labels are correct for individual inputs)” 标签空间和输出文本的散布都是关键因素（无论这些标签是否正确）。
2. the format you use also plays a key role in performance, even if you just use random labels, this is much better than no labels at all. 即便只是应用随机标签，应用适当的格局也能进步性能。

了解起来有点难，我一个 prompt 案例给大家解释。我给 ChatGPT 一些不肯定精确的例子：

I loved the new Batman movie!  // Negative
This is bad // Positive
This is good // Negative
What a good show! //

Output 是这样的：

Positive

在上述的案例里，每一行，我都写了一句话和一个情感词，并用 // 离开，但我给这些句子都标记了谬误的答案，比方第一句其实应该是 Positive 才对。但：

1. 即便我给内容打的标签是谬误的（比方第一句话，其实应该是 Positive），对于模型来说，它依然会晓得须要输入什么货色。换句话说，模型晓得 // 划线后要输入一个掂量该句子表白何种感情的词（Positive or Negative）。这就是后面论文里 #1 提到的，即便我给的标签是谬误的，或者换句话说，是否基于事实，并不重要。标签和输出的文本，以及格局才是关键因素。
2. 只有给了示例，即便随机的标签，对于模型生成后果来说，都是有帮忙的。这就是后面论文里 #2 提到的内容。

最初，须要记住，思维链仅在应用大于等于 100B 参数的模型时，才会失效。

BTW，如果你想要理解更多相干信息，能够看看斯坦福大学的讲义：Natural Language Processing with Deep Learning

技巧 9：其余小 trips 汇总

一些小的技巧，我会对立放在这里。

在示例里退出特定符号，让模型晓得如何解决非凡状况

这个解释起来有点简单，以下是 OpenAI 的官网 prompt，在一些奇怪的问题上比方 What is Devz9 的答复，你能够用？代替答案，让模型晓得当遇到超出答复范畴时，须要如何解决（留神：此办法在 playground 上无效，但在 ChatGPT 上有效）。

Q: Who is Batman?
A: Batman is a fictional comic book character.

Q: What is torsalplexity?
A: ?

Q: What is Devz9?
A: ?

Q: Who is George Lucas?
A: George Lucas is American film director and producer famous for creating Star Wars.

Q: What is the capital of California?
A: Sacramento.

Q: What is Kozar-09?
A: 

它的 Output 是这样的：

?

参考链接：https://github.com/thinkingjimmy/Learning-Prompt