RAG高級優化：檢索后處理模塊成竹在胸
作者：哎呀AIYA
根據論文 Lost in the Middle: How Language Models Use Long Contexts，的實驗表明，大模型更容易記憶開頭和結尾的文檔，而對中間部分的文檔記憶能力不強，因此可以根據召回的文檔和query的相關性進行重排序。
本文我們將介紹在將召回片段送入大模型之前的一些優化手段，它們能幫助大模型更好的理解上下文知識，給出最佳的回答：
- Long-text Reorder
- Contextual compression
- Refine
- Emotion Prompt
Long-text Reorder
根據論文 Lost in the Middle: How Language Models Use Long Contexts，的實驗表明，大模型更容易記憶開頭和結尾的文檔，而對中間部分的文檔記憶能力不強，因此可以根據召回的文檔和query的相關性進行重排序。
圖片
核心的代碼可以參考langchain的實現：
def _litm_reordering(documents: List[Document]) -> List[Document]:
"""Lost in the middle reorder: the less relevant documents will be at the
middle of the list and more relevant elements at beginning / end.
See: https://arxiv.org/abs//2307.03172"""
documents.reverse()
reordered_result = []
for i, value in enumerate(documents):
if i % 2 == 1:
reordered_result.append(value)
else:
reordered_result.insert(0, value)
return reordered_resultContextual compression
本質上利用LLM去判斷檢索之后的文檔和用戶query的相關性，只返回相關度最高的k個。
from langchain.retrievers import ContextualCompressionRetriever
from langchain.retrievers.document_compressors import LLMChainExtractor
from langchain_openai import OpenAI
llm = OpenAI(temperature=0)
compressor = LLMChainExtractor.from_llm(llm)
compression_retriever = ContextualCompressionRetriever(
base_compressor=compressor, base_retriever=retriever
)
compressed_docs = compression_retriever.get_relevant_documents(
"What did the president say about Ketanji Jackson Brown"
)
print(compressed_docs)Refine
對最后大模型生成的回答進行進一步的改寫，保證回答的準確性。主要涉及提示詞工程，參考的提示詞如下：
The original query is as follows: {query_str}
We have provided an existing answer: {existing_answer}
We have the opportunity to refine the existing answer (only if needed) with some more context below.
------------
{context_msg}
------------
Given the new context, refine the original answer to better answer the query. If the context isn't useful, return the original answer.
Refined Answer:Emotion Prompt
同樣是提示詞工程的一部分，思路來源于微軟的論文：
Large Language Models Understand and Can Be Enhanced by Emotional Stimuli
在論文中，微軟研究員提出，在提示詞中增加一些情緒情感相關的提示，有助于大模型輸出高質量的回答。
參考提示詞如下：
emotion_stimuli_dict = {
"ep01": "Write your answer and give me a confidence score between 0-1 for your answer. ",
"ep02": "This is very important to my career. ",
"ep03": "You'd better be sure.",
# add more from the paper here!!
}
# NOTE: ep06 is the combination of ep01, ep02, ep03
emotion_stimuli_dict["ep06"] = (
emotion_stimuli_dict["ep01"]
+ emotion_stimuli_dict["ep02"]
+ emotion_stimuli_dict["ep03"]
)
from llama_index.prompts import PromptTemplate
qa_tmpl_str = """\
Context information is below.
---------------------
{context_str}
---------------------
Given the context information and not prior knowledge, \
answer the query.
{emotion_str}
Query: {query_str}
Answer: \
"""
qa_tmpl = PromptTemplate(qa_tmpl_str)責任編輯：武曉燕
來源：
哎呀AIYA
