Lossy LLM

Weekly I/O#55

ChatGPT is like a lossy compression of all the text on the Web. If it takes material generated by itself as training data for next model, the output will be worse. Can LLM's output be as good as its own input?

Article: ChatGPT Is a Blurry JPEG of the Web

This might not be factually true but I found the analogy interesting. In Ted Chiang's words:

"Think of ChatGPT as a blurry jpeg of all the text on the Web. It retains much of the information on the Web, in the same way that a jpeg retains much of the information of a higher-resolution image, but, if you’re looking for an exact sequence of bits, you won’t find it; all you will ever get is an approximation. But, because the approximation is presented in the form of grammatical text, which ChatGPT excels at creating, it’s usually acceptable. You’re still looking at a blurry jpeg, but the blurriness occurs in a way that doesn’t make the picture as a whole look less sharp"

"But I’m going to make a prediction: when assembling the vast amount of text used to train GPT-4, the people at OpenAI will have made every effort to exclude material generated by ChatGPT or any other large-language model. If this turns out to be the case, it will serve as unintentional confirmation that the analogy between large-language models and lossy compression is useful. Repeatedly resaving a creates more compression artifacts, because more information is lost every time. It’s the digital equivalent of repeatedly making photocopies of photocopies in the old days. The image quality only gets worse."

"Indeed, a useful criterion for gauging a large-language model’s quality might be the willingness of a company to use the text that it generates as training material for a new model. If the output of ChatGPT isn’t good enough for GPT-4, we might take that as an indicator that it’s not good enough for us, either. Conversely, if a model starts generating text so good that it can be used to train new models, then that should give us confidence in the quality of that text. (I suspect that such an outcome would require a major breakthrough in the techniques used to build these models.) If and when we start seeing models producing output that’s as good as their input, then the analogy of lossy compression will no longer be applicable."

Want to learn 5 bite-sized cool things like this every week to understand the world better? Sign up below for my free weekly newsletter and learn together!

Weeklyio Banner