考拉的Ai树屋

Appearance

校正RAG（CRAG） ​


如需转载，请联系微信群主

校正RAG（CRAG）是RAG的一种策略，它在检索到的文档上结合了自反射/自分级。

在此论文中here，采取了以下几个步骤：

  • 如果至少有一个文档的相关性超过了阈值，则继续生成
  • 在生成之前，它会进行知识精炼
  • 这会将文档划分为“知识条”
  • 它对每个知识条进行评分，并过滤掉不相关的部分
  • 如果所有文档的相关性都低于阈值或评分者不确定，则框架会寻找额外的数据源
  • 它将使用网页搜索来补充检索

我们将使用 LangGraph 从头实现这些想法：

  • 作为第一次尝试，我们将跳过知识精炼阶段。如果需要，可以将其添加回一个节点。
  • 如果任何文档不相关，我们将选择通过网页搜索来补充检索。
  • 我们将使用 Tavily Search 进行网页搜索。
  • 让我们使用查询重写来优化网页搜索的查询。

环境 ​

! pip install langchain_community tiktoken langchain-openai langchainhub chromadb langchain langgraph tavily-python

LLMs

python
import os

os.environ["OPENAI_API_KEY"] = ""

搜索

我们将使用 Tavily Search 进行网页搜索。

python
os.environ["TAVILY_API_KEY"] = ""

追踪

可选地，通过设置使用 LangSmith 进行追踪（示例在底部显示）。

python
os.environ["LANGCHAIN_TRACING_V2"] = "true"
os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"
os.environ["LANGCHAIN_API_KEY"] = ""

索引

让我们索引三篇博客文章。

python
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

urls = [
    "https://lilianweng.github.io/posts/2023-06-23-agent/",
    "https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/",
    "https://lilianweng.github.io/posts/2023-10-25-adv-attack-llm/",
]

docs = [WebBaseLoader(url).load() for url in urls]
docs_list = [item for sublist in docs for item in sublist]

text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
    chunk_size=250, chunk_overlap=0
)
doc_splits = text_splitter.split_documents(docs_list)

# Add to vectorDB
vectorstore = Chroma.from_documents(
    documents=doc_splits,
    collection_name="rag-chroma",
    embedding=OpenAIEmbeddings(),
)
retriever = vectorstore.as_retriever()

LLMs ​

python
### Retrieval Grader

from langchain_core.prompts import ChatPromptTemplate
from langchain_core.pydantic_v1 import BaseModel, Field
from langchain_openai import ChatOpenAI


# Data model
class GradeDocuments(BaseModel):
    """Binary score for relevance check on retrieved documents."""

    binary_score: str = Field(
        description="Documents are relevant to the question, 'yes' or 'no'"
    )


# LLM with function call
llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)
structured_llm_grader = llm.with_structured_output(GradeDocuments)

# Prompt
system = """You are a grader assessing relevance of a retrieved document to a user question. \n 
    If the document contains keyword(s) or semantic meaning related to the question, grade it as relevant. \n
    Give a binary score 'yes' or 'no' score to indicate whether the document is relevant to the question."""
grade_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        ("human", "Retrieved document: \n\n {document} \n\n User question: {question}"),
    ]
)

retrieval_grader = grade_prompt | structured_llm_grader
question = "agent memory"
docs = retriever.get_relevant_documents(question)
doc_txt = docs[1].page_content
print(retrieval_grader.invoke({"question": question, "document": doc_txt}))
binary_score='yes'
python
### Generate

from langchain import hub
from langchain_core.output_parsers import StrOutputParser

# Prompt
prompt = hub.pull("rlm/rag-prompt")

# LLM
llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0)


# Post-processing
def format_docs(docs):
    return "\n\n".join(doc.page_content for doc in docs)


# Chain
rag_chain = prompt | llm | StrOutputParser()

# Run
generation = rag_chain.invoke({"context": docs, "question": question})
print(generation)
The design of generative agents combines LLM with memory, planning, and reflection mechanisms to enable agents to behave conditioned on past experience. Memory stream is a long-term memory module that records a comprehensive list of agents' experience in natural language. Short-term memory is utilized for in-context learning, while long-term memory allows agents to retain and recall information over extended periods.
python
### Question Re-writer

# LLM
llm = ChatOpenAI(model="gpt-3.5-turbo-0125", temperature=0)

# Prompt
system = """You a question re-writer that converts an input question to a better version that is optimized \n 
     for web search. Look at the input and try to reason about the underlying semantic intent / meaning."""
re_write_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        (
            "human",
            "Here is the initial question: \n\n {question} \n Formulate an improved question.",
        ),
    ]
)

question_rewriter = re_write_prompt | llm | StrOutputParser()
question_rewriter.invoke({"question": question})
'What is the role of memory in artificial intelligence agents?'

网络搜索工具 ​

python
### Search

from langchain_community.tools.tavily_search import TavilySearchResults

web_search_tool = TavilySearchResults(k=3)

图 ​

以图方式捕捉流程。

图的状态 ​

python
from typing import List

from typing_extensions import TypedDict


class GraphState(TypedDict):
    """
    Represents the state of our graph.

    Attributes:
        question: question
        generation: LLM generation
        web_search: whether to add search
        documents: list of documents
    """

    question: str
    generation: str
    web_search: str
    documents: List[str]
python
from langchain.schema import Document


def retrieve(state):
    """
    Retrieve documents

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): New key added to state, documents, that contains retrieved documents
    """
    print("---RETRIEVE---")
    question = state["question"]

    # Retrieval
    documents = retriever.get_relevant_documents(question)
    return {"documents": documents, "question": question}


def generate(state):
    """
    Generate answer

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): New key added to state, generation, that contains LLM generation
    """
    print("---GENERATE---")
    question = state["question"]
    documents = state["documents"]

    # RAG generation
    generation = rag_chain.invoke({"context": documents, "question": question})
    return {"documents": documents, "question": question, "generation": generation}


def grade_documents(state):
    """
    Determines whether the retrieved documents are relevant to the question.

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): Updates documents key with only filtered relevant documents
    """

    print("---CHECK DOCUMENT RELEVANCE TO QUESTION---")
    question = state["question"]
    documents = state["documents"]

    # Score each doc
    filtered_docs = []
    web_search = "No"
    for d in documents:
        score = retrieval_grader.invoke(
            {"question": question, "document": d.page_content}
        )
        grade = score.binary_score
        if grade == "yes":
            print("---GRADE: DOCUMENT RELEVANT---")
            filtered_docs.append(d)
        else:
            print("---GRADE: DOCUMENT NOT RELEVANT---")
            web_search = "Yes"
            continue
    return {"documents": filtered_docs, "question": question, "web_search": web_search}


def transform_query(state):
    """
    Transform the query to produce a better question.

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): Updates question key with a re-phrased question
    """

    print("---TRANSFORM QUERY---")
    question = state["question"]
    documents = state["documents"]

    # Re-write question
    better_question = question_rewriter.invoke({"question": question})
    return {"documents": documents, "question": better_question}


def web_search(state):
    """
    Web search based on the re-phrased question.

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): Updates documents key with appended web results
    """

    print("---WEB SEARCH---")
    question = state["question"]
    documents = state["documents"]

    # Web search
    docs = web_search_tool.invoke({"query": question})
    web_results = "\n".join([d["content"] for d in docs])
    web_results = Document(page_content=web_results)
    documents.append(web_results)

    return {"documents": documents, "question": question}


### Edges


def decide_to_generate(state):
    """
    Determines whether to generate an answer, or re-generate a question.

    Args:
        state (dict): The current graph state

    Returns:
        str: Binary decision for next node to call
    """

    print("---ASSESS GRADED DOCUMENTS---")
    state["question"]
    web_search = state["web_search"]
    state["documents"]

    if web_search == "Yes":
        # All documents have been filtered check_relevance
        # We will re-generate a new query
        print(
            "---DECISION: ALL DOCUMENTS ARE NOT RELEVANT TO QUESTION, TRANSFORM QUERY---"
        )
        return "transform_query"
    else:
        # We have relevant documents, so generate answer
        print("---DECISION: GENERATE---")
        return "generate"

构建图

这只是按照我们上面概述的流程进行的。

python
from langgraph.graph import END, StateGraph

workflow = StateGraph(GraphState)

# Define the nodes
workflow.add_node("retrieve", retrieve)  # retrieve
workflow.add_node("grade_documents", grade_documents)  # grade documents
workflow.add_node("generate", generate)  # generatae
workflow.add_node("transform_query", transform_query)  # transform_query
workflow.add_node("web_search_node", web_search)  # web search

# Build graph
workflow.set_entry_point("retrieve")
workflow.add_edge("retrieve", "grade_documents")
workflow.add_conditional_edges(
    "grade_documents",
    decide_to_generate,
    {
        "transform_query": "transform_query",
        "generate": "generate",
    },
)
workflow.add_edge("transform_query", "web_search_node")
workflow.add_edge("web_search_node", "generate")
workflow.add_edge("generate", END)

# Compile
app = workflow.compile()
python
from pprint import pprint

# Run
inputs = {"question": "What are the types of agent memory?"}
for output in app.stream(inputs):
    for key, value in output.items():
        # Node
        pprint(f"Node '{key}':")
        # Optional: print full state at each node
        # pprint.pprint(value["keys"], indent=2, width=80, depth=None)
    pprint("\n---\n")

# Final generation
pprint(value["generation"])
---RETRIEVE---
"Node 'retrieve':"
'\n---\n'
---CHECK DOCUMENT RELEVANCE TO QUESTION---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
"Node 'grade_documents':"
'\n---\n'
---ASSESS GRADED DOCUMENTS---
---DECISION: ALL DOCUMENTS ARE NOT RELEVANT TO QUESTION, TRANSFORM QUERY---
---TRANSFORM QUERY---
"Node 'transform_query':"
'\n---\n'
---WEB SEARCH---
"Node 'web_search_node':"
'\n---\n'
---GENERATE---
"Node 'generate':"
'\n---\n'
"Node '__end__':"
'\n---\n'
('Agents possess short-term memory, which is utilized for in-context learning, '
 'and long-term memory, allowing them to retain and recall vast amounts of '
 'information over extended periods. Some experts also classify working memory '
 'as a distinct type, although it can be considered a part of short-term '
 'memory in many cases.')
python
from pprint import pprint

# Run
inputs = {"question": "How does the AlphaCodium paper work?"}
for output in app.stream(inputs):
    for key, value in output.items():
        # Node
        pprint(f"Node '{key}':")
        # Optional: print full state at each node
        # pprint.pprint(value["keys"], indent=2, width=80, depth=None)
    pprint("\n---\n")

# Final generation
pprint(value["generation"])
---RETRIEVE---
"Node 'retrieve':"
'\n---\n'
---CHECK DOCUMENT RELEVANCE TO QUESTION---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
"Node 'grade_documents':"
'\n---\n'
---ASSESS GRADED DOCUMENTS---
---DECISION: ALL DOCUMENTS ARE NOT RELEVANT TO QUESTION, TRANSFORM QUERY---
---TRANSFORM QUERY---
"Node 'transform_query':"
'\n---\n'
---WEB SEARCH---
"Node 'web_search_node':"
'\n---\n'
---GENERATE---
"Node 'generate':"
'\n---\n'
"Node '__end__':"
'\n---\n'
('The AlphaCodium paper functions by proposing a code-oriented iterative flow '
 'that involves repeatedly running and fixing generated code against '
 'input-output tests. Its key mechanisms include generating additional data '
 'like problem reflection and test reasoning to aid the iterative process, as '
 'well as enriching the code generation process. AlphaCodium aims to improve '
 'the performance of Large Language Models on code problems by following a '
 'test-based, multi-stage approach.')

如需转载，请联系微信群主

加群：

扫描下方二维码加好友，添加申请填写“ai加群”，成功添加后，回复“ai加群”或耐心等待管理员邀请你入群