4 ms·

The question I have is whether the tools to moderate ChatGPT and correct its' wrong answers should be in place for humans anyway. It's not like human workers ar

by yunwal 4y ago

The question I have is whether the tools to moderate ChatGPT and correct its' wrong answers should be in place for humans anyway. It's not like human workers are 100% reliable processes, and in some cases we scale human work to dangerous levels. Ultimately, the best way to make sure an answer is correct is to come to it from multiple directions. If we use GPT and other AI models as another direction it seems like a strict win to me.

pixl97 4y ago
Robert Miles recently did a video on this and even that may not be enough. This appears to be a really hard problem. https://www.youtube.com/watch?v=w65p_IIp6JY https://www.youtube.com/watch?v=w65p_IIp6JY
- yunwal 4y ago
  Hmm, by multiple directions, what I really meant is, consult a model trained in 2 vastly different ways, or consult a human + a model, or consult a model + a calculator, or a model + human-written test cases. He didn't really go into the idea of "if 2 independent processes can arrive at the same factual statement, it's probably the truth" angle here. Of course, 2 independent processes is still a tough thing to design when you require this much data, and the data is probably coming from web-scraping, so I think for now, human-generated test cases are still needed.
  - pixl97 4y ago
    I mean factual independence likely works for math and repeatable science, but this starts to break down because a lot of interesting things are historical and determining actual independence is very difficult. For example the openAI model and people working for openAI submitting are not what I consider independent. This goes the same for any human involved processes. For example if you're talking about a religion the most vocal, and most likely for input are most likely to be very for or very against it, and no rational middle ground may even exist.
    - yunwal 4y ago
      The context of this discussion was a coding question: would you use LLMs as a coding assistant? In general, this method works for things where there's an objective desired result, which includes most coding applications.

by yunwal 4y ago

pixl97 4y ago

Robert Miles recently did a video on this and even that may not be enough. This appears to be a really hard problem. https://www.youtube.com/watch?v=w65p_IIp6JY https://www.youtube.com/watch?v=w65p_IIp6JY

yunwal 4y ago
Hmm, by multiple directions, what I really meant is, consult a model trained in 2 vastly different ways, or consult a human + a model, or consult a model + a calculator, or a model + human-written test cases. He didn't really go into the idea of "if 2 independent processes can arrive at the same factual statement, it's probably the truth" angle here. Of course, 2 independent processes is still a tough thing to design when you require this much data, and the data is probably coming from web-scraping, so I think for now, human-generated test cases are still needed.
- pixl97 4y ago
  I mean factual independence likely works for math and repeatable science, but this starts to break down because a lot of interesting things are historical and determining actual independence is very difficult. For example the openAI model and people working for openAI submitting are not what I consider independent. This goes the same for any human involved processes. For example if you're talking about a religion the most vocal, and most likely for input are most likely to be very for or very against it, and no rational middle ground may even exist.
  - yunwal 4y ago
    The context of this discussion was a coding question: would you use LLMs as a coding assistant? In general, this method works for things where there's an objective desired result, which includes most coding applications.