Comment by firefoxd
1 month ago
If you are serious about sharing written ideas, I suggest you avoid using this type of prompts at all cost. I've worked with LLMs to write on my blog and they are pretty good at first glance [0]. But do it a few time and you'll notice that those tropes are the least of your problems. Not only all your articles will sound the same, but you'll see that same voice on other blogs, news articles, white paper, etc. It's as if they were all written by Mo Samuels. Readers are often here for the author's voice, not just the content of the text.
I often hear this here: "if you don't bother writing, why should I bother reading?" In fact, save us some time and just share the prompt.
I have seen people suggest that the problem is that LLMs let you express any of your ideas, but the number of people with ideas worth expressing is limited.
In a sense I think this is accurate, but not inevitable. I think there is a lack of creative thinking, but it has come from a world that doesn't value it and suppresses difference.
There is a brilliant line in Treehouse of Horrors IV where Principle Skinner says "Now I've gotten word that a child is using his imagination, and I've come to put a stop to it." Which is just the perfect comment on the modern education system.
Models trained on the lack of diversity will push one way, but I think it will also avenues for expression that didnb't exist before. The balance will come from how we react and support what we would like to have happen
I think it has more to do with LLM's being statistical models than human creativity lacking in the input. The creativity and millions of voices and tones may be there, but since these models tend to go for the most likely next words, polishing this away becomes a feature.
A text by a human mind may be seen as a jagged crystal with rough edges and character. Maybe not perfectly written but it's special.
An LLM takes a million of crystals and trims the most likely tokens to be chosen into what would rather appear as a smooth pebble; the common core of all crystals. And everyone using the LLM will get very similar pebbles because to the LLM, regardless who is speaking to it, it will provide the same most likely next tokens. It's not that creativity is lacking in the input, but the LLM picks the most commonly chosen words by all humans in given contexts.
For that to sound imaginative and great as you go, it would have to not only exist in the data, but be a common dominating voice among humans. But if it was, it wouldn't be seen as creative because it would be the new normal.
So I'm not sure how there's a good way out of this. You could push LLM temperature high so that it becomes more "creative" by picking less popular tokens as it writes, but this instead tend to make it unpredictable and picking words it shouldn't have. I mean, we are still dealing with statistical models here rather than brains and it's a rough tool for that job.
>I think it has more to do with LLM's being statistical models than human creativity lacking in the input. The creativity and millions of voices and tones may be there, but since these models tend to go for the most likely next words, polishing this away becomes a feature.
I have always thought this is a rather misguided view as to what LLMs do and indeed what statistical models are. When people describe something as 'just statistics' I feel like they have a rather high-school-ish view of what statistics represents and are transferring this simplistic view to what is going on inside a LLM. Notably they do not find the most probable next word. They find the probability of every word that could come next. That is a far richer signal than most imagine.
And ultimately it's like saying that human brains are just chemical bonds changing and sometimes triggering electrical pulses that causes some more chemicals to change. Complex arrangements of simple mechanisms can produce human thought. Pointing at any simple internal mechanism of an entity without taking into account the structural complexity would force you to assume that both AI and Humans are incapable of creativity.
Transformers are essentially multi-layer perceptron with a mechanism attached to transfer information to where it is needed.
3 replies →
> I have seen people suggest that the problem is that LLMs let you express any of your ideas, but the number of people with ideas worth expressing is limited.
It doesn't just have to be one problem.
1. Laundering your "ideas" through an LLM makes them less of your ideas, at best you get the classic two sentences of content embedded in two pages of padding.
2. LLMs removed a filter that help cut down on the amount of useless writing we'd have to wade through. The difficulty of expressing an idea acts as a filter to weed out many (but not all) ideas not worth expressing. That applies to both to people with ideas worth expressing and those without.
On the former, I've had the experience of having an idea, then witnessing it fall apart as I try to express it, as I think about it more deeply. LLMs let you avoid that.
> I often hear this here: "if you don't bother writing, why should I bother reading?"
That is an opinion somebody shared on X which has been mindlessly repeated over and over again in other places such as this site.
Why do you value those comments when all they are doing is parroting something they didn’t think themselves? It seems to undermine your point entirely. There is zero originality or effort in those comments. Why are you bothering to read them?
Copying and pasting somebody else’s opinion from one social media site to another is no more virtuous than what you are complaining about.
I value that opinion because it resonates with me. When I use an LLM to write an article, it's usually because I don't have the time or energy to go through my normal process of writing.
Sure, I still end up with a polished article, but a lot of it is not entirely my idea or something I would have written through the filter of my own experience. So in order to share my true take on a subject, I have to go through the struggle of writing and bouncing of ideas in my head, which almost always results in a better output.
"Sharing the prompt" is a category error. It assumes the value of a piece is in the instructions given to the model, rather than the proprietary input or the iterative editing that follows. There is a hard line between using an LLM to generate content from a void and using it to synthesize specific ideas.
If someone asks a model to "write a post about X," they are outsourcing the thinking, which results in the homogenized voice everyone is tired of.