Nice work ! It still manage to use the word 'delve' in the first sentence, which is a giveaway that it's written by a LLM.
[0] many “LLM tells” fit this pattern of just being common features of professionally-published works that are less often seen in casual writing.
LLMs are raising the bar by expanding the vocabulary people are exposed to, so words like delve will stick out. I think it's preferred by writers because it articulates a nice sounding alternative to words like explore, venture, analyze, think about, confront, etc. It's a useful, versatile word, and one of the metrics by which writers measure quality is the minimization of syllables.
LLMs are mostly indistinguishable from humans at this point; a one-shot output from any of the major models can be recognized in the same way you might recognize a writer. With multiple style passes, you're not going to be able to tell the difference between ChatGPT, Ronald Reagan, Bill Clinton, Hunter S. Thompson, Einstein, or any other sufficiently modeled figure. Throw in a few tens of thousands of words written by yourself and most of the models will do a nearly flawless job of copying your stylometric profile.
Language communicates ideas, and we've made machines that produce intricate, sophisticated ideas that land in our brains. The consequences are going to be fascinating.
It would be totally nuts to take points off for using spell check. An LLM should be able to provide style check without causing any concerns; it will become the norm, and then too good prose won’t cause any flags to be thrown.
> It was into that bank that the creature had worked its way, and on listening I could hear it delving and scraping at a great rate, about a yard from the back of the wall.
I bring that up to point out that this isn't necessarily (more) common in the 19th century style print literature, so the observation might not be silly. The model creating the modern synthetic version injected 'delve' 9 times, which implies that it is more frequently used in modern literature or just something that models tend to inject. Though, I could be missing something (either in searching the data set, or how this works).
[1] https://huggingface.co/datasets/dleemiller/irish_penny_journ...
I think we're well beyond the point where the majority of people cannot tell what is actually produced by an LLM and we're convincing ourselves we still have a handle on the situation. A lot of these rules are completely arbitrary and vary from person to person.
RIP kids who grew up digesting hundreds of fantasy novels and playing D&D.
If anything, I'd assume the chatbots would use Indian English phrases like "do the needful" and those weird phrases that only make sense in Hindi but are translated to English.
It's used a ton by LLMs for some reason despite being rarely used by real people. I think it's mostly a byproduct of LLMs having their training data being over represented by certain published works instead of casual communications.
There does seem to be something else going on with delve specifically though, one of the other comments mentions that delve isn't used in the specific training data for this, so it's odd to see it being used in the output. I wonder if it's because delve has secondary definitions of "to make a careful or detailed search for information" and "to examine a subject in detail" which is causing the LLM to use it to seem like it's answers are more thorough.
The popular theory is that it's due to English-language RLHF tasks being outsourced to Nigeria, where "delve" is used relatively often.