O recuperador com reconhecimento de histórico discutido nesta postagem é aquele retornado pela função create_history_aware_retriever do pacote LangChain. Esta função foi projetada para receber as seguintes entradas em seu construtor:

• Um LLM (um modelo de linguagem que recebe uma consulta e retorna uma resposta);
• Um recuperador de armazenamento de vetores (um modelo que recebe uma consulta e retorna uma lista de documentos relevantes).
• Um histórico de bate-papo (uma lista de interações de mensagens, normalmente entre um ser humano e uma IA).

Quando invocado, o recuperador com reconhecimento de histórico recebe uma consulta do usuário como entrada e gera uma lista de documentos relevantes. Os documentos relevantes são baseados na consulta combinada com o contexto fornecido pelo histórico do chat.

No final, resumi seu fluxo de trabalho.

Configurando

from langchain.chains import create_history_aware_retriever
from langchain_community.document_loaders import WebBaseLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_chroma import Chroma
from dotenv import load_dotenv
import bs4

load_dotenv() # To get OPENAI_API_KEY

def create_vectorsore_retriever():
    """
    Returns a vector store retriever based on the text of a specific web page.
    """
    URL = r'https://lilianweng.github.io/posts/2023-06-23-agent/'
    loader = WebBaseLoader(
        web_paths=(URL,),
        bs_kwargs=dict(
            parse_only=bs4.SoupStrainer(class_=("post-content", "post-title", "post-header"))
        ))
    docs = loader.load()
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0, add_start_index=True)
    splits = text_splitter.split_documents(docs)
    vectorstore = Chroma.from_documents(documents=splits, embedding=OpenAIEmbeddings())
    return vectorstore.as_retriever()

def create_prompt():
    """
    Returns a prompt instructed to produce a rephrased question based on the user's
    last question, but referencing previous messages (chat history).
    """
    system_instruction = """Given a chat history and the latest user question \
        which might reference context in the chat history, formulate a standalone question \
        which can be understood without the chat history. Do NOT answer the question, \
        just reformulate it if needed and otherwise return it as is."""

    prompt = ChatPromptTemplate.from_messages([
        ("system", system_instruction),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}")])
    return prompt

llm = ChatOpenAI(model='gpt-4o-mini')
vectorstore_retriever = create_vectorsore_retriever()
prompt = create_prompt()

history_aware_retriever = create_history_aware_retriever(
    llm,
    vectorstore_retriever,
    prompt
)

Usando

Aqui, uma pergunta é feita sem nenhum histórico de bate-papo, então o recuperador responde apenas com os documentos relevantes para a última pergunta.

chat_history = []

docs = history_aware_retriever.invoke({'input': 'what is planning?', 'chat_history': chat_history})
for i, doc in enumerate(docs):
    print(f'Chunk {i 1}:')
    print(doc.page_content)
    print()

Chunk 1:
Planning is essentially in order to optimize believability at the moment vs in time.
Prompt template: {Intro of an agent X}. Here is X's plan today in broad strokes: 1)
Relationships between agents and observations of one agent by another are all taken into consideration for planning and reacting.
Environment information is present in a tree structure.

Chunk 2:
language. Essentially, the planning step is outsourced to an external tool, assuming the availability of domain-specific PDDL and a suitable planner which is common in certain robotic setups but not in many other domains.

Chunk 3:
Another quite distinct approach, LLM P (Liu et al. 2023), involves relying on an external classical planner to do long-horizon planning. This approach utilizes the Planning Domain Definition Language (PDDL) as an intermediate interface to describe the planning problem. In this process, LLM (1) translates the problem into “Problem PDDL”, then (2) requests a classical planner to generate a PDDL plan based on an existing “Domain PDDL”, and finally (3) translates the PDDL plan back into natural

Chunk 4:
Planning

Subgoal and decomposition: The agent breaks down large tasks into smaller, manageable subgoals, enabling efficient handling of complex tasks.
Reflection and refinement: The agent can do self-criticism and self-reflection over past actions, learn from mistakes and refine them for future steps, thereby improving the quality of final results.


Memory

Agora, com base no histórico do bate-papo, o recuperador sabe que o humano deseja saber sobre a decomposição de tarefas e também sobre o planejamento. Portanto, ele responde com pedaços de texto que fazem referência a ambos os temas.

chat_history = [
    ('human', 'when I ask about planning I want to know about Task Decomposition too.')]

docs = history_aware_retriever.invoke({'input': 'what is planning?', 'chat_history': chat_history})
for i, doc in enumerate(docs):
    print(f'Chunk {i 1}:')
    print(doc.page_content)
    print()

Chunk 1:
Task decomposition can be done (1) by LLM with simple prompting like "Steps for XYZ.\n1.", "What are the subgoals for achieving XYZ?", (2) by using task-specific instructions; e.g. "Write a story outline." for writing a novel, or (3) with human inputs.

Chunk 2:
Fig. 1. Overview of a LLM-powered autonomous agent system.
Component One: Planning#
A complicated task usually involves many steps. An agent needs to know what they are and plan ahead.
Task Decomposition#

Chunk 3:
Planning

Subgoal and decomposition: The agent breaks down large tasks into smaller, manageable subgoals, enabling efficient handling of complex tasks.
Reflection and refinement: The agent can do self-criticism and self-reflection over past actions, learn from mistakes and refine them for future steps, thereby improving the quality of final results.


Memory

Chunk 4:
Challenges in long-term planning and task decomposition: Planning over a lengthy history and effectively exploring the solution space remain challenging. LLMs struggle to adjust plans when faced with unexpected errors, making them less robust compared to humans who learn from trial and error.

Agora, toda a questão é baseada no histórico de bate-papo. E podemos ver que ele responde com pedaços de texto que fazem referência ao conceito correto.

chat_history = [
    ('human', 'What is ReAct?'),
    ('ai', 'ReAct integrates reasoning and acting within LLM by extending the action space to be a combination of task-specific discrete actions and the language space')]

docs = history_aware_retriever.invoke({'input': 'It is a way of doing what?', 'chat_history': chat_history})
for i, doc in enumerate(docs):
    print(f'Chunk {i 1}:')
    print(doc.page_content)
    print()

Chunk 1:

ReAct (Yao et al. 2023) integrates reasoning and acting within LLM by extending the action space to be a combination of task-specific discrete actions and the language space. The former enables LLM to interact with the environment (e.g. use Wikipedia search API), while the latter prompting LLM to generate reasoning traces in natural language.

The ReAct prompt template incorporates explicit steps for LLM to think, roughly formatted as:

Thought: ...

Action: ...

Observation: ...

Chunk 2:

Fig. 2.  Examples of reasoning trajectories for knowledge-intensive tasks (e.g. HotpotQA, FEVER) and decision-making tasks (e.g. AlfWorld Env, WebShop). (Image source: Yao et al. 2023).

In both experiments on knowledge-intensive tasks and decision-making tasks, ReAct works better than the Act-only baseline where Thought: … step is removed.

Chunk 3:

The LLM is provided with a list of tool names, descriptions of their utility, and details about the expected input/output.

It is then instructed to answer a user-given prompt using the tools provided when necessary. The instruction suggests the model to follow the ReAct format - Thought, Action, Action Input, Observation.

Chunk 4:

Case Studies#

Scientific Discovery Agent#

ChemCrow (Bran et al. 2023) is a domain-specific example in which LLM is augmented with 13 expert-designed tools to accomplish tasks across organic synthesis, drug discovery, and materials design. The workflow, implemented in LangChain, reflects what was previously described in the ReAct and MRKLs and combines CoT reasoning with tools relevant to the tasks:

Conclusão

Concluindo, o fluxo de trabalho dos recuperadores com reconhecimento de histórico funciona da seguinte maneira quando .invoke({'input': '...', 'chat_history': '...'}) é chamado:

• Ele substitui os espaços reservados input e chat_history no prompt por valores especificados, criando um novo prompt pronto para uso que essencialmente diz "Pegue este histórico de bate-papo e esta última entrada e reformule a última entrada de uma forma que qualquer pessoa possa entender sem ver o histórico do chat".
• Ele envia o novo prompt para o LLM e recebe uma entrada reformulada.
• Em seguida, ele envia a entrada reformulada para o recuperador de armazenamento de vetores e recebe uma lista de documentos relevantes para essa entrada reformulada.
• Finalmente, ele retorna esta lista de documentos relevantes.

Obs.: É importante ressaltar que o embedding utilizado para transformar texto em vetor é aquele especificado quando Chroma.from_documents é chamado. Quando nenhum é especificado (o presente caso), a incorporação de croma padrão é usada.