--- title: A Preliminary Exploration into Factored Cognition with... description: EleutherAI researchers investigate factored cognition — breaking complex tasks into smaller subtasks — as a way to improve GPT-3's performance on arithmetic... canonical: https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: A Preliminary Exploration into Factored Cognition with Language Models | daily.dev og:description: EleutherAI researchers investigate factored cognition — breaking complex tasks into smaller subtasks — as a way to improve GPT-3's performance on arithmetic... og:url: https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x og:image: https://api.daily.dev/og/posts/rVmkn1l3x.png og:image:alt: A Preliminary Exploration into Factored Cognition with Language Models 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. # A Preliminary Exploration into Factored Cognition with Language Models **[EleutherAI](https://daily.dev/sources/eleutherai)** · 7 min read · 0 upvotes · 0 comments ## Summary EleutherAI researchers investigate factored cognition — breaking complex tasks into smaller subtasks — as a way to improve GPT-3's performance on arithmetic and math problems it cannot otherwise solve. Experiments cover nested expression evaluation (linear decomposition), branched nested function evaluation (exponential branching), and open-ended math word problems from MathQA. Results show that stepwise decomposition significantly outperforms direct answering, though accuracy degrades rapidly as task depth and branching increase. The MathQA experiments reveal GPT-3 still struggles with noisy, conceptually demanding word problems even with rationale-based few-shot prompting. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [The cost of thinking](https://daily.dev/posts/the-cost-of-thinking-zi8tusdea) · MIT News · 0 upvotes · 0 comments - [AI models get better at math but still get low marks](https://daily.dev/posts/ai-models-get-better-at-math-but-still-get-low-marks-bugdv2bcl) · The Register · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#nlp](https://daily.dev/tags/nlp) [View this post on daily.dev](https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x) ```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":"A Preliminary Exploration into Factored Cognition with Language Models","url":"https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x"},"datePublished":"2026-08-02T07:04:29.235Z","dateModified":"2026-08-02T07:31:15.682Z","description":"EleutherAI researchers investigate factored cognition — breaking complex tasks into smaller subtasks — as a way to improve GPT-3's performance on arithmetic...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/7fa3fbd6e99da7e05ab42115e7e68fde?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/7fa3fbd6e99da7e05ab42115e7e68fde?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"EleutherAI","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":"EleutherAI","logo":"https://media.daily.dev/image/upload/s--aAcRyWGF--/f_auto,q_auto/v1785654263/logos/eleutherai","url":"https://daily.dev/sources/eleutherai"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/a-preliminary-exploration-into-factored-cognition-with-language-models-rvmkn1l3x","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,nlp","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"EleutherAI","item":"https://daily.dev/sources/eleutherai"},{"@type":"ListItem","position":3,"name":"A Preliminary Exploration into Factored Cognition with Language Models"}]} ```