Read our paper on de-slopping LLM outputs. It's far more than simply all having the same fixed AI system prompts. It's an overuse of post-training and contempt for pre-training.
You're supposed to also remove the fancy UTF-8 quotes that people can't normally type, the EM dashes, and reorder sentences/clauses because the paragraph level "template" slop is really obvious to people who use these models all the time. (I'm also pretty sure that the UTF-8 shenanigans with LLM responses was done very on purpose by those who have a vested interest in making it easier for mass surveillance of written communication.)
Or, use the "deep research" mode for writing your prose instead. It's far less sloppy in how it writes.
These people are amateurs at humanizing their writing.
We wrote the paper on how to deslop your LLM outputs and if you use our factory de-slopped versions of gemma3 you don't have to worry about this, similarly if you use our antislop sampler, your LLM outputs will look very close to human.
You better check your privilege re: your oppressive claims about humans needing to live somewhere to avoid trauma. Do you think the Roma have a lot of trauma just because they are nomads?
Nomadic people have homes and possessions; they bring the homes with them. They don't live out in the open with nothing like unhoused people are compelled to do.
There are digital nomads too - they usually have money and live in rented places, but they have shelter.
Powerwash simulator is occasionally fun. There's shiny rewards, I don't have to deal with potential bad weather, and there's no random patches that take 20 times to get rid of. If I don't feel like powerwashing simulator, it will wait for me, forever, with no ill consequences or social judgement.
If I never wash my actual driveway, the same is not true. Therefore I will need to wash it at times when it's unpleasant or I don't want to, and it will take longer than powerwashing a driveway in Powerwash simulator.
Unfortunately you’re right re:dating prospects but that’s mostly because game devs haven’t been able to reproduce the insane success of valorant at getting the better gender to want to play hardcore games.
Ding ding ding ding, we have another person who actually understands how to use this feature.
The fact that most people don't know any of these things that you are mentioning is one of the myriad reasons why the most killer feature of LLMs continues to languish in obscurity.
BTW, the structured outputs debate is significantly more complicated than even your own post implies.
You aren't testing structured outputs+model alone, you are testing
1. The structured outputs backend used. There are at least 3 major free ones, outlines, xgrammer, lm-format-enforcer and guidance. OpenAI, Anthropic, Google, and Grok will all have different ones. They all do things SIGNIFICANTLY differently. That's at least 8 different backends to compare.
2. The settings used for each structured output backend. Oh, you didn't know that there's often 5+ settings related to how they handle subtle stuff like whitespaces? Better learn to figure out what these settings do and how to tweak them!
3. The models underlying sampling settings, i.e. any default temperature, top_p/top_k, etc going on. Remember that the ORDER of application of samplers matters here! Huggingface transformers and vLLM have opposite defaults on if temperature happens before sampling or after!
4. The model, and don't forget about differences around quants/variants of the model!
Almost no one who does any kinds of these analysis even talk about these additional factors, including academics.
Sometimes it feels like I'm the only one in this world who actually uses this feature at the extremes of its capabilities.
https://arxiv.org/pdf/2510.15061
reply