– by Alexander Kovalev
LlamaIndex
LlamaIndex is a project that provides a central interface to connect your LLM’s with external data like database and files(pdf, json, docx, epub etc). It also can access to some site like bilibili, reddit etc. There’s a Llama Hub to collect all the interfaces.
LlamaIndex provides a simple pattern to connect your external data to OpenAI:
- Build
documentby usingReader - Create
indexfromdocument - Query with your questions by using
index.qeury
Simple like that. More cases could be found in its Ref
简单来说，LlamaIndex可以看成是一个将外部数据接入LLM的连接器。有了LlamaIndex，用户可以将存在于各种静态文本或者动态网站APP的数据接入LLM。 由于有了统一的使用模式，所以无论什么外部数据源，大家都可以将其导入成Document对象，然后通过Index来对数据提问。底层它接入了OpenAI的API，所以需要OpenAI API的token (不小心导入了一本神雕侠侣和一本笑傲江湖，终于成功的把免费的token用完了)。
LlamaIndex 还为各种接口提供了Hub，所以如果想要什么导入器，可以先去Hub上找找
目前这个社区相当活跃，大家都在贡献各种数据源，甚至包括B站。
LangChain
这个是相对更加偏模型侧的项目，通过引入Agents, Prompts, Chains 等把prompt分解成更细的prompt和行为。和LlamaIndex一样，LangChain 也提供了一个 Hub站点，供其他人去找优质的Prompt。
The core idea of the library is that we can “chain” together different components to create more advanced use cases around LLMs. Chains may consist of multiple components from several modules:
- Prompt templates: Prompt templates are templates for different types of prompts. Like “chatbot” style templates, ELI5 question-answering, etc
- LLMs: Large language models like GPT-3, BLOOM, etc
- Agents: Agents use LLMs to decide what actions should be taken. Tools like web search or calculators can be used, and all are packaged into a logical loop of operations.
- Memory: Short-term memory, long-term memory.
- Chain: Chain allow us to combine multiple components together to create a single, coherent application. For example,
LLMChainis a simple chain that takes in a prompt template, formats it with the user input and returns the response from an LLM.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
from langchain.prompts import PromptTemplate
from langchain.llms import OpenAI
from langchain.chains import LLMChain
llm = OpenAI(temperature=0.9)
prompt = PromptTemplate(
input_variables=["product"],
template="What is a good name for a company that makes {product}?",
)
chain = LLMChain(llm=llm, prompt=prompt)
# Run the chain only specifying the input variable.
print(chain.run("colorful socks"))
(AI)对每个人都将产生深远和系统性影响。我们的假设是每个人很快将有副驾驶员，不光是1个，可能是5个、6个。 有些副驾驶员足够强，变成正驾驶员，他自动可以去帮你做事。更长期，我们每个人都有一个驾驶员团队服务。未来的人类组织是真人，加上他的副驾驶员和真驾驶员一起协同。 – 陆奇
Key concepts
Chain of Thoughts
Chain of Thoughts (CoT) is a prompting technique used to encourage the model to generate a series of intermediate reasoning steps. A less formal way to induce this behavior is to include “Let’s think step-by-step” in the prompt.
简单来说，就是由你来告诉大模型一个问题的中间推理步骤，一步步引导得出问题的最终答案。 它其实是Few-Shot Prompt的一种形式。
这里需要解释一下什么叫Few-Shot Prompt，它是相对于Zero-Shot Prompt来说的。 Zero-Shot Prompt 就是那种针对问题直接给答案的提示语，没有任何辅助信息的提问。 这种提问基本上就是基于模型的预训练数据，所以有时候得到的准确性偏低。
为了提高答案的准确性，比较推荐的是 Few-Shot Prompt，就是在提问中给模型一些例子，比如想让他给你任何一个词的反义词，那么你可以先给它几个例子比如说高的反义词是矮， 内的反义词是外等等，之后再把你想要问的词输给它，那么它的答复准确性能提高很多。
Here is a sample of CoT Here is a sample of how to use chain of thoughts to solve a problem:
I am hungry. I need to eat something. I could make a sandwich. I need bread, meat, and cheese. I have bread in the fridge. I need to go to the store for meat and cheese. I will go to the store after I finish this work.
Action Plan Generation
Action Plan Generation is a prompting technique that uses a language model to generate actions to take. The results of these actions can then be fed back into the language model to generate a subsequent action.
行动计划生成是指创建一个实现特定目标的行动计划的过程。它涉及识别需要采取的步骤、所需资源和完成的时间表。
ReAct
ReAct is a prompting technique that combines Chain-of-Thought prompting with action plan generation. This induces the to model to think about what action to take, then take it.