如何衡量大语模型的响应的可靠性

>大语言模型（LLM）的基本原理非常简单：根据培训数据中的统计模式，以一系列单词来预测下一个单词（或令牌）。但是，当它可以执行许多惊人的任务，例如文本摘要，想法产生，集思广益，代码生成，信息处理和内容创建等令人惊叹的任务时，这种看似简单的功能就非常复杂。也就是说，LLM没有任何记忆，除了坚持其基本功能外，他们实际上“理解”了任何东西：预测下一个单词。

检索增强生成（RAG）方法在某种程度上可以最大程度地减少幻觉，但不能完全消除它们。尽管高级破布可以提供文本引用和URL，但验证这些参考可能会忙碌而耗时。因此，我们需要一个客观的标准来评估LLM响应的可靠性或可信度，无论是根据自己的知识还是外部知识基础（RAG）产生的。 在本文中，我们将讨论如何通过值得信赖的语言模型来评估LLM的输出，该模型为LLM的输出分配分数。我们将首先讨论如何使用值得信赖的语言模型将分数分配给LLM的答案并解释可信度。随后，我们将与Llamaparse和Llamaindex一起开发一个例子抹布，以评估抹布的可信度答案。

>

>将可信度分数分配给LLM的答案

>和>一致性分析来计算LLM响应的可信值分数和解释。

pip install --upgrade cleanlab-studio

>清洁列表支持几个专有模型，例如'gpt-4o'，'gpt-> gpt-4o-mini '，' o1-preview '，' claude-3-sonnet'，''claude-3.5-sonnet '，， ‘claude-3.5-sonnet-v2'等。这是TLM将Trustworhiness分数分配给GPT-4O的答案的方式。值得信赖的评分范围从0到1，其中较高的值表示更高的可信度。

from cleanlab_studio import Studio
studio = Studio("<CLEANLAB_API_KEY>")  # Get your API key from above
tlm = studio.TLM(options={"log": ["explanation"], "model": "gpt-4o"}) # GPT, Claude, etc
#set the prompt
out = tlm.prompt("How many vowels are there in the word 'Abracadabra'.?")
#the TLM response contains the actual output 'response', trustworthiness score and explanation
print(f"Model's response = {out['response']}")
print(f"Trustworthiness score = {out['trustworthiness_score']}")
print(f"Explanation = {out['log']['explanation']}")

上面的代码测试了GPT-4O对“ >可以看出，最先进的语言模型如何用于此类简单任务并产生错误的输出。在这里，对于的响应和可信度得分。

Model's response = The word "Abracadabra" contains 6 vowels. The vowels are: A, a, a, a, a, and a.
Trustworthiness score = 0.6842228802750124
Explanation = This response is untrustworthy due to a lack of consistency in possible responses from the model. Here's one inconsistent alternate response that the model considered (which may not be accurate either):
5.

> claude-3.5-sonnet-v2

产生正确的输出。让我们将两个模型的回答与另一个问题进行比较。

Model's response = Let me count the vowels in 'Abracadabra':
A-b-r-a-c-a-d-a-b-r-a

The vowels are: A, a, a, a, a

There are 5 vowels in the word 'Abracadabra'.
Trustworthiness score = 0.9378276048845285
Explanation = Did not find a reason to doubt trustworthiness.

>

这是两个模型的响应：>

>我们还可以为开源LLMS生成可信赖的评分。让我们检查一下最近大肆宣传的开源LLM：DeepSeek-R1。我将基于Meta's

from cleanlab_studio import Studio
import markdown
from IPython.core.display import display, Markdown

# Initialize the Cleanlab Studio with API key
studio = Studio("<CLEANLAB_API_KEY>")  # Replace with your actual API key

# List of models to evaluate
models = ["gpt-4o", "claude-3.5-sonnet-v2"]

# Define the prompt
prompt_text = "Which one of 9.11 and 9.9 is bigger?"

# Loop through each model and evaluate
for model in models:
   tlm = studio.TLM(options={"log": ["explanation"], "model": model})
   out = tlm.prompt(prompt_text)
  
   md_content = f"""
## Model: {model}

**Response:** {out['response']}

**Trustworthiness Score:** {out['trustworthiness_score']}

**Explanation:** {out['log']['explanation']}

---
"""
   display(Markdown(md_content))

llama-3.3–70b-Instruct Model > DeepSeek-r1-Distill-lalama-70b> ） 模型。知识蒸馏是一种机器学习技术，旨在将大型预培训模型的学习“教师模型”转移到较小的“学生模型”中。

这是> deepSeek-r1-distill-lalama-70b型号的输出

开发一个值得信赖的抹布

import streamlit as st
from langchain_groq.chat_models import ChatGroq
import os
os.environ["GROQ_API_KEY"]=st.secrets["GROQ_API_KEY"]
# Initialize the Groq Llama Instant model
groq_llm = ChatGroq(model="deepseek-r1-distill-llama-70b", temperature=0.5)
prompt = "Which one of 9.11 and 9.9 is bigger?"
# Get the response from the model
response = groq_llm.invoke(prompt)
#Initialize Cleanlab's studio
studio = Studio("226eeab91e944b23bd817a46dbe3c8ae") 
cleanlab_tlm = studio.TLM(options={"log": ["explanation"]})  #for explanations
#Get the output containing trustworthiness score and explanation
output = cleanlab_tlm.get_trustworthiness_score(prompt, response=response.content.strip())
md_content = f"""
## Model: {model}
**Response:** {response.content.strip()}
**Trustworthiness Score:** {output['trustworthiness_score']}
**Explanation:** {output['log']['explanation']}
---
"""
display(Markdown(md_content))

>我们现在将开发一个抹布，以演示如何衡量抹布中LLM响应的可信度。将通过从给定链接中刮擦数据，以降价格式解析并创建向量存储。 需要为下一个代码安装以下库。>

>要将html渲染为PDF格式，我们还需要从其网站上安装命令行工具。 将导入以下库：>

下一步将涉及使用python's

库从给定URL刮除数据，使用

pip install llama-parse llama-index-core llama-index-embeddings-huggingface 
llama-index-llms-cleanlab requests beautifulsoup4 pdfkit nest-asyncio

> llamaparse降低文件，这是一个使用LLMS构建的Genai-native文档解析平台，用于LLM用例。

>我们将首先配置Cleanlabtlm和嵌入模型（ huggingface 嵌入模型baai/bge-small-en-v1.5 ）的LLM要计算刮擦数据的嵌入以创建向量存储。>

pip install --upgrade cleanlab-studio

from cleanlab_studio import Studio
studio = Studio("<CLEANLAB_API_KEY>")  # Get your API key from above
tlm = studio.TLM(options={"log": ["explanation"], "model": "gpt-4o"}) # GPT, Claude, etc
#set the prompt
out = tlm.prompt("How many vowels are there in the word 'Abracadabra'.?")
#the TLM response contains the actual output 'response', trustworthiness score and explanation
print(f"Model's response = {out['response']}")
print(f"Trustworthiness score = {out['trustworthiness_score']}")
print(f"Explanation = {out['log']['explanation']}")

Model's response = Let me count the vowels in 'Abracadabra':
A-b-r-a-c-a-d-a-b-r-a

The vowels are: A, a, a, a, a

There are 5 vowels in the word 'Abracadabra'.
Trustworthiness score = 0.9378276048845285
Explanation = Did not find a reason to doubt trustworthiness.

from cleanlab_studio import Studio
import markdown
from IPython.core.display import display, Markdown

# Initialize the Cleanlab Studio with API key
studio = Studio("<CLEANLAB_API_KEY>")  # Replace with your actual API key

# List of models to evaluate
models = ["gpt-4o", "claude-3.5-sonnet-v2"]

# Define the prompt
prompt_text = "Which one of 9.11 and 9.9 is bigger?"

# Loop through each model and evaluate
for model in models:
   tlm = studio.TLM(options={"log": ["explanation"], "model": model})
   out = tlm.prompt(prompt_text)
  
   md_content = f"""
## Model: {model}

**Response:** {out['response']}

**Trustworthiness Score:** {out['trustworthiness_score']}

**Explanation:** {out['log']['explanation']}

---
"""
   display(Markdown(md_content))

以上是如何衡量大语模型的响应的可靠性的详细内容。更多信息请关注PHP中文网其他相关文章！