--- title: Train LLMs better, without using language...? | daily.dev description: A new AI research paper proposes pre-training language models on synthetic worlds generated by neural cellular automata before exposing them to natural... canonical: https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Train LLMs better, without using language...? | daily.dev og:description: A new AI research paper proposes pre-training language models on synthetic worlds generated by neural cellular automata before exposing them to natural... og:url: https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj og:image: https://api.daily.dev/og/posts/mu25mqJAJ.png og:image:alt: Train LLMs better, without using language...? og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Train LLMs better, without using language...? **[bycloud](https://daily.dev/sources/bycloud)** · 1 min read · 0 upvotes · 0 comments ## Summary A new AI research paper proposes pre-training language models on synthetic worlds generated by neural cellular automata before exposing them to natural language. By learning to infer hidden rules from evolving patterns (similar to Conway's Game of Life), models develop fundamental skills like pattern tracking and dependency understanding. Training on just 164 million tokens of this synthetic data improved subsequent language training by ~6% and made it up to 1.6x faster, outperforming models pre-trained on significantly more natural text. The finding suggests that teaching abstract inference before language leads to better and faster language learning. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Training Language Models via Neural Cellular Automata](https://daily.dev/posts/training-language-models-via-neural-cellular-automata-sg0rcdmbq) · Hacker News · 2 upvotes · 0 comments - [Meet the new biologists treating LLMs like aliens](https://daily.dev/posts/meet-the-new-biologists-treating-llms-like-aliens-i8raaa1kt) · MIT Technology Review · 1 upvotes · 0 comments - [How LLMs Learn from the Internet: The Training Process](https://daily.dev/posts/how-llms-learn-from-the-internet-the-training-process-i9yw0ueye) · ByteByteGo · 2 upvotes · 0 comments - [Researchers discover a shortcoming that makes LLMs less reliable](https://daily.dev/posts/researchers-discover-a-shortcoming-that-makes-llms-less-reliable-wiiiapmcn) · MIT News · 3 upvotes · 0 comments --- [View this post on daily.dev](https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Train LLMs better, without using language...?","url":"https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj"},"datePublished":"2026-04-03T18:14:48.066Z","dateModified":"2026-04-03T18:15:27.777Z","description":"A new AI research paper proposes pre-training language models on synthetic worlds generated by neural cellular automata before exposing them to natural...","image":"https://i.ytimg.com/vi/bncKD7P4VoU/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/bncKD7P4VoU/sddefault.jpg","isAccessibleForFree":true,"articleSection":"bycloud","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"bycloud","logo":"https://media.daily.dev/image/upload/s--UUgAibgb--/f_auto/v1712315122/logos/bycloud","url":"https://daily.dev/sources/bycloud"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/train-llms-better-without-using-language--mu25mqjaj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"","timeRequired":"PT1M","video":{"@type":"VideoObject","name":"Train LLMs better, without using language...?","description":"A new AI research paper proposes pre-training language models on synthetic worlds generated by neural cellular automata before exposing them to natural...","thumbnailUrl":"https://i.ytimg.com/vi/bncKD7P4VoU/sddefault.jpg","uploadDate":"2026-04-03T18:14:48.066Z","duration":"PT1M","url":"https://api.daily.dev/r/mu25mqJAJ","embedUrl":"https://www.youtube.com/embed/bncKD7P4VoU"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"bycloud","item":"https://daily.dev/sources/bycloud"},{"@type":"ListItem","position":3,"name":"Train LLMs better, without using language...?"}]} ```