Antislop: A Comprehensive Framework for Identifying and Eliminating Repetitive Patterns in Language Models

This is a better way than we use to create the "slop ban list" in our "Antislop" sampler/fine-tuning method. Gonna totally nab the weights for all these words and remove this crap from claude going forward! Thanks OP!!! https://arxiv.org/abs/2510.15061 (ICLR 2026)

Der_EinzigeAug 28, 2026View on HNParent context

Oh, sorry about that: https://arxiv.org/abs/2510.15061 (ICLR 2026) (not actually sorry)

Der_EinzigeAug 10, 2026View on HNParent context

For anyone that wants to evade the kind of people who want to figure out if an AI wrote your review or not, we wrote a whole paper (ICLR 2026!) on how to do that! https://arxiv.org/abs/2510.15061 I consider all types of "I liked this output, but don't the moment I learned it was…

Der_EinzigeJul 31, 2026View on HNParent context

things like that indeed tend to be fraught some relevant links https://arxiv.org/abs/2408.02442 https://arxiv.org/abs/2510.15061 https://arxiv.org/abs/2604.13006

tensegristJul 14, 2026View on HNParent context

We send our regards: https://arxiv.org/abs/2510.15061 (ICLR 2026)

Der_EinzigeJun 30, 2026View on HNParent context

We solved this problem already with the antislop sampler: https://arxiv.org/abs/2510.15061

Der_EinzigeJun 1, 2026View on HNParent context

Related, we made "Antislop", a framework for removing repetitive phraseology from LLMs: https://arxiv.org/abs/2510.15061

Der_EinzigeMay 29, 2026View on HNParent context

We figured out how to remove that crap in our ICLR 2026 paper: https://arxiv.org/pdf/2510.15061

Der_EinzigeApr 1, 2026View on HNParent context

We wrote the paper on deslopping LLM and their outputs: https://arxiv.org/abs/2510.15061

Der_EinzigeFeb 19, 2026View on HNParent context

FWIW: When I accuse folks of it, take it more seriously - we literally wrote a paper (ICLR 2026) on deslopping language models: https://arxiv.org/abs/2510.15061 Also you might not care, but normies sure care! https://arxiv.org/html/2410.03723v2 https://arxiv.org/html/2510.08831v1

Der_EinzigeFeb 16, 2026View on HNParent context

This is what they get for not reading our antislop paper (ICLR 2026) and using our anti-slopped sampler/models, or Kimi (which is remarkable relatively non sloppy) https://arxiv.org/abs/2510.15061 I thought normies would have caught onto the EM dash, overuse of semicolons…

Der_EinzigeFeb 13, 2026View on HNParent context

Just use our antislop techniques and no one will ever know you used an LLM. https://arxiv.org/abs/2510.15061 (ICLR 2026) Also you could long use "logit_bias" in the API of models which supported it to ban the EM dash, ban the word "not", ban semicolons, and ban the "fancy…

Der_EinzigeFeb 12, 2026View on HNParent context

Our paper on removing AI slop got accepted to ICLR 2026, and it's under consideration for an IgNobel prize: https://arxiv.org/abs/2510.15061 Our definition of slop (repetitive characteristic language from LLMs) is the original one as articulated by the LLM creative writing…

Der_EinzigeFeb 8, 2026View on HNParent context

This still doesn't remove all the slop. You need sampler or fine-tuning tricks for it. https://arxiv.org/abs/2510.15061

Der_EinzigeFeb 6, 2026View on HNParent context

We wrote the actual paper on “stopping slop” https://arxiv.org/abs/2510.15061

Der_EinzigeJan 21, 2026View on HNParent context

Go take a giant dataset of LLM generated outputs, use an accurate POS tagger and look for 5-grams or similar lengths of matching patterns. If you do thi, you’ll pull out the overrepresented paragraph and sentence level slop that we humans intuitively detect easily. If your…

Der_EinzigeJan 20, 2026View on HNParent context

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. https://arxiv.org/abs/2510.15061

Der_EinzigeDec 27, 2025View on HNParent context

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. https://arxiv.org/pdf/2510.15061

Der_EinzigeDec 27, 2025View on HNParent context

We wrote the paper on how to remove slop from LLMs. https://arxiv.org/abs/2510.15061 Also somewhat tangentially relevant video: https://www.youtube.com/watch?v=Tsp2bC0Db8o

Der_EinzigeNov 26, 2025View on HNParent context

And people like you will hate it even more when the normies immunize themselves from being obviously caught by such tells: https://arxiv.org/abs/2510.15061

Der_EinzigeNov 14, 2025View on HNParent context

You can remove this easily. We wrote a paper on how to remove this slop from LLMs. https://arxiv.org/abs/2510.15061

Der_EinzigeNov 14, 2025View on HNParent context

We wrote the paper on how to deslop your language model: https://arxiv.org/abs/2510.15061

Der_EinzigeNov 13, 2025View on HNParent context

Pretty sad that they didn't even try our antislop sampler. Slop in LLM outputs is a choice: https://arxiv.org/abs/2510.15061

Der_EinzigeNov 6, 2025View on HNParent context

ScholarlyArticle: "Antislop: A Comprehensive Framework for Identifying and Eliminating Repetitive Patterns in Language Models" (2025) https://arxiv.org/abs/2510.15061 : > Abstract: [...] Our approach combines three innovations: (1) The Antislop Sampler, which uses backtracking…

westurnerOct 23, 2025View on HNParent context