From: derek@shape-of-code.com   
      
   John,   
      
   > [I would like to see some actual data. In my experience, LLMs are   
   > impressive, confident, and frequently wrong. -John]   
      
   LLM's performance on fact recall is poor.   
      
   It seems to be much better than other tools when dealing with   
   grammars. I could not find the example I was looking for, but here are   
   two others: https://arxiv.org/abs/2305.19234   
   https://szopa.medium.com/teaching-chatgpt-to-speak-my-sons-inven   
   ed-language-9d109c0a0f05   
      
   My own experience using local (i.e., very small)   
   models   
   https://shape-of-code.com/2024/02/25/extracting-named-entities-f   
   om-a-change-log-using-an-llm/   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)