<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <link href="https://quaily.com/aiexperimentsthoughts/feed/atom" rel="self" type="application/atom+xml" />
  <title><![CDATA[Interesting AI Experiments &Thoughts]]></title>
  <subtitle type="html"><![CDATA[Share some interesting experiments, thoughts and papers in AI]]></subtitle>
  <updated>2024-07-23T07:06:46Z</updated>
  <author>
    <name>chengshengdeng</name>
  </author>
  
  <logo>https://static.quail.ink/media/r8zplhdde7.webp</logo>
  <icon>https://static.quail.ink/media/r8zplhdde7.webp</icon>
  <id>https://quaily.com/aiexperimentsthoughts</id>
  <generator uri="https://quaily.com" version="1.0">Quaily</generator>

  
  <entry>
    <title><![CDATA[July 23, DSPy with GPT-4o-mini on MMLU-Pro]]></title>
    <link href="https://quaily.com/aiexperimentsthoughts/p/july-23-dspy-with-gpt-4o-mini-on-mmlu-pro" />
    <id>https://quaily.com/aiexperimentsthoughts/p/july-23-dspy-with-gpt-4o-mini-on-mmlu-pro#4791</id>
    <author>
      <name>chengshengdeng</name>
    </author>
    <published>2024-07-23T07:06:46Z</published>
    <updated>2024-07-23T07:06:46Z</updated>
    <summary>
      <![CDATA[ DSPy is an optimization framework that enhances prompts and responses from models like GPT-4o-mini. It showcases the magic of the framework and demonstra...ers. ]]>
    </summary>
    <content type="html">
      <![CDATA[  ]]>
    </content>
  </entry>
  
  <entry>
    <title><![CDATA[July 16, 2024 LLMs Evals Thoughts]]></title>
    <link href="https://quaily.com/aiexperimentsthoughts/p/july-16-2024-llms-evals-thoughts" />
    <id>https://quaily.com/aiexperimentsthoughts/p/july-16-2024-llms-evals-thoughts#4561</id>
    <author>
      <name>chengshengdeng</name>
    </author>
    <published>2024-07-16T15:36:08Z</published>
    <updated>2024-07-16T15:36:08Z</updated>
    <summary>
      <![CDATA[ Evaluating LLMs is important for understanding their abilities and solving real business problems. A good evaluation requires sufficient and high-quality...ime. ]]>
    </summary>
    <content type="html">
      <![CDATA[  ]]>
    </content>
  </entry>
  
  <entry>
    <title><![CDATA[July 14, 2024 How to use Yi-Vision with TextGrad]]></title>
    <link href="https://quaily.com/aiexperimentsthoughts/p/july-15-2024-how-to-use-yi-vision-with-textgrad" />
    <id>https://quaily.com/aiexperimentsthoughts/p/july-15-2024-how-to-use-yi-vision-with-textgrad#4508</id>
    <author>
      <name>chengshengdeng</name>
    </author>
    <published>2024-07-14T09:50:20Z</published>
    <updated>2024-07-14T09:50:20Z</updated>
    <summary>
      <![CDATA[ TextGrad is an autograd engine that enhances language models through iterative feedback. It has recently expanded to support multimodal optimization. Thi...wer. ]]>
    </summary>
    <content type="html">
      <![CDATA[  ]]>
    </content>
  </entry>
  
  <entry>
    <title><![CDATA[July 9, How to use DeepSeek with TextGrad]]></title>
    <link href="https://quaily.com/aiexperimentsthoughts/p/july-9-how-to-use-deepseek-with-textgrad" />
    <id>https://quaily.com/aiexperimentsthoughts/p/july-9-how-to-use-deepseek-with-textgrad#4444</id>
    <author>
      <name>chengshengdeng</name>
    </author>
    <published>2024-07-09T07:57:13Z</published>
    <updated>2024-07-09T07:57:13Z</updated>
    <summary>
      <![CDATA[ TextGrad can be used with models like DeepSeek not just OpenAI, allowing for optimization. ]]>
    </summary>
    <content type="html">
      <![CDATA[  ]]>
    </content>
  </entry>
  
  <entry>
    <title><![CDATA[July, 2024 LLMs Evaluation Benchmarks]]></title>
    <link href="https://quaily.com/aiexperimentsthoughts/p/july-2024-llms-evaluation-benchmarks" />
    <id>https://quaily.com/aiexperimentsthoughts/p/july-2024-llms-evaluation-benchmarks#4403</id>
    <author>
      <name>chengshengdeng</name>
    </author>
    <published>2024-07-06T11:00:54Z</published>
    <updated>2024-07-06T11:00:54Z</updated>
    <summary>
      <![CDATA[ Evaluation benchmarks for Large Language Models (LLMs) are being updated to match their evolving capabilities. This blog explores several commonly refere...ore. ]]>
    </summary>
    <content type="html">
      <![CDATA[  ]]>
    </content>
  </entry>
  
</feed>
