Do LLMs write like humans? Variation in grammatical and rhetorical styles · HackerPapers discussions

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Thu, 29 Oct 2026 20:04:10 GMT

The "regression to the mean" or "it's just the training data" hypothesis doesn't work. Pretrained models do not have these weird writing styles and sound much more human. It's only after instruction tuning -- SFT, DPO, etc. -- that LLMs develop weird writing styles. That is, the original training on massive text corpora seems to produce an average of human writing, but that average looks pretty average. It doesn't feel like LLM slop. But the instruction tuning on specific tasks (answer this prompt, summarize this text, do this math problem), which turns text predictors into chatbots, seems to…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Wed, 23 Sep 2026 22:49:45 GMT

It's not the training data. When we've tested the writing style of base models that have not gone through instruction tuning, they're much more human-like (https://arxiv.org/abs/2410.16107). The style shift seems to come from something in the instruction-tuning process, so our current research problem is figuring out what in the process is doing it.

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Mon, 10 Aug 2026 22:43:26 GMT

> Well, what do you expect? LLMs are trained on blithering, mostly from web sites. So you get blithering out. I realize this isn't entirely serious, but I can't resist pointing out that this doesn't seem to be a good explanation for why LLMs write the way they do. When we've experimented with LLM writing style on open-weights models where you can get a base model (pretraining on text only) and an instruction-tuned variant (pretraining + post-training with RLHF and whatever other human-evaluated tasks), it's the instruction-tuned variant that shows the weird writing quirks. That is, the…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — HN submission

Mon, 15 Jun 2026 00:32:43 GMT

Show HN: Isitslop.xyz, a free AI-writing detector built on latest research

I have a weird habit of guessing whether something's AI before I check, especially on LinkedIn and Reddit (for obvious reasons). The detectors I used to confirm were mostly junk and built to sell you a "humanizer" upsell, so I went and read the actual research on what gives AI text away in 2026, then built my own (no login and free) The main sources I used are these: https://arxiv.org/abs/2406.07016 https://arxiv.org/abs/2403.07183 https://aclanthology.org/2025.coling-main.426/ https://arxiv.org/abs/2410.16107 https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing (Amazing article on the…

5 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Sat, 07 Mar 2026 23:46:07 GMT

I work on research studying LLM writing styles, so I am going to have to steal this. I've seen plenty of lists of LLM style features, but this is the first one I noticed that mentions "tapestry", which we found is GPT-4o's second-most-overused word (after "camaraderie", for some reason).[1] We used a set of grammatical features in our initial style comparisons (like present participles, which GPT-4o loved so much that they were a pretty accurate classifier on their own), but it shouldn't be too hard to pattern-match some of these other features and quantify them. If anyone who works on LLMs…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Wed, 27 Aug 2025 22:41:54 GMT

Sure, if you're learning to write and want lots of examples of a particular style, LLMs can generate that for you. Just don't assume that is a normal writing style, or that it matches a particular genre (say, workplace communication, or academic writing, or whatever). Our experience (https://arxiv.org/abs/2410.16107) is that LLMs like GPT-4o have a particular writing style, including both vocabulary and distinct grammatical features, regardless of the type of text they're prompted with. The style is informationally dense, features longer words, and favors certain grammatical structures (like…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Thu, 17 Jul 2025 23:45:59 GMT

I don't think the AI companies are systematically working to make their models sound more human. They're working to make them better at specific tasks, but the writing styles are, if anything, even more strange as they advance. Comparing base and instruction-tuned models, the base models are vaguely human in style, while instruction-tuned models systematically prefer certain types of grammar and style features. (For example, GPT-4o loves participial clauses and nominalizations.) https://arxiv.org/abs/2410.16107 When I've looked at more recent models like o3, there are other style shifts. The…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Mon, 07 Jul 2025 22:51:16 GMT

In our studies of ChatGPT's grammatical style (https://arxiv.org/abs/2410.16107), it really loves past and present participial phrases (2-5x more usage than humans). I didn't see any here in a glance through the lightfastness section, though I didn't try running the whole article through spaCy to check. In any case it doesn't trip my mental ChatGPT detector either; it reads more like classic SEO writing you'd see all over blogs in the 20-teens. edit: yeah, ran it through our style feature tagger and nothing jumps out. Low rate of nominalizations (ChatGPT loves those), only a few present…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Fri, 28 Feb 2025 12:36:28 GMT

Maybe, but I'm not sure how much the style is deliberate vs. a consequence of the post-training tasks like summarization and problem solving. Without seeing the post-training tasks and rating systems it's hard to judge if it's a deliberate style or an emergent consequence of other things. But it's definitely the case that base models sound more human than instruction-tuned variants. And the shift isn't just vocabulary, it's also in grammar and rhetorical style. There's a shift toward longer words, but also participial phrases, phrasal coordination (with "and" and "or"), and nominalizations…

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by capnrefsmmat

Tue, 25 Feb 2025 17:56:30 GMT

Freely available preprint: https://arxiv.org/abs/2410.16107 I'm biased (as an author), but I think Figure 3 is the key result, as well as Tables 4 and 5 in the appendix. (Or Table 1 in the published version.) The grammatical and rhetorical differences in the models, even when asked to match an example of human text, are remarkably big. They quantify that uncanny feeling you get when encoutering LLM-written text in the wild.

Paper and discussion history. A mention is not an endorsement.

Do LLMs write like humans? Variation in grammatical and rhetorical styles — mentioned by nomat

Tue, 17 Dec 2024 23:53:50 GMT

the authors of this paper[0] were able to use AI to figure out if a block of text was written by AI, so hopefully this becomes a tool available to educators very soon. [0] https://arxiv.org/html/2410.16107v1

Paper and discussion history. A mention is not an endorsement.