--- title: Google Research just proved you can boost llm accuracy... description: Google Research published findings showing that simply duplicating a prompt (pasting it twice) can improve LLM accuracy by up to 76 percentage points, with no... canonical: https://daily.dev/posts/google-research-just-proved-you-can-boost-llm-accuracy-by-up-to-76-percentage-points-with-zero-extra-2a4odqkz8 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Google Research just proved you can boost llm accuracy by up to 76 percentage points with zero extra output tokens, zero latency increase, and zero fine-tuning 🤯 the technique: paste your prompt twice. that's it. that's the paper. but WHY it works reveals something important about how every llm you use actually reads your input: | daily.dev og:description: Google Research published findings showing that simply duplicating a prompt (pasting it twice) can improve LLM accuracy by up to 76 percentage points, with no... og:url: https://daily.dev/posts/google-research-just-proved-you-can-boost-llm-accuracy-by-up-to-76-percentage-points-with-zero-extra-2a4odqkz8 og:image: https://api.daily.dev/og/posts/2A4oDQKZ8.png og:image:alt: Google Research just proved you can boost llm accuracy by up to 76 percentage points with zero extra output tokens, zero latency increase, and zero fine-tuning 🤯 the technique: paste your prompt twice. that's it. that's the paper. but WHY it works reveals something important about how every llm you use actually reads your input: og:image:width: 1200 og:image:height: 630 og:locale: en --- [Robert Youssef](https://daily.dev/sources/rryssf%5F) [Read on](https://api.daily.dev/r/2A4oDQKZ8) Feb 24 • From x.com  Robert Youssef @rryssf\_ Google Research just proved you can boost llm accuracy by up to 76 percentage points with zero extra output tokens, zero latency increase, and zero fine-tuning 🤯 the technique: paste your prompt twice. that's it. that's the paper. but WHY it works reveals something important about how every llm you use actually reads your input:  Comment Bookmark Copy Share your thoughts Post Share this post [](https://daily.dev/sources/rryssf%5F) [Robert Youssef](https://daily.dev/sources/rryssf%5F "https://daily.dev/sources/rryssf_") 0 Followers • 6 Upvotes #### Would you recommend this post? Copy link Slack WhatsApp Facebook X New Squad Copy linkShare with your friends ```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":"Google Research just proved you can boost llm accuracy by up to 76 percentage points with zero extra output tokens, zero latency increase, and zero fine-tuning 🤯\n\nthe technique: paste your prompt twice.\n\nthat's it. that's the paper.\n\nbut WHY it works reveals something important about how every llm you use actually reads your input:","url":"https://daily.dev/posts/google-research-just-proved-you-can-boost-llm-accuracy-by-up-to-76-percentage-points-with-zero-extra-2a4odqkz8","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/google-research-just-proved-you-can-boost-llm-accuracy-by-up-to-76-percentage-points-with-zero-extra-2a4odqkz8"},"datePublished":"2026-02-24T16:45:34.351Z","dateModified":"2026-02-26T09:41:24.743Z","description":"Google Research published findings showing that simply duplicating a prompt (pasting it twice) can improve LLM accuracy by up to 76 percentage points, with no...","image":"https://pbs.twimg.com/media/HB7_Y9iaMAIcgXD.jpg","thumbnailUrl":"https://pbs.twimg.com/media/HB7_Y9iaMAIcgXD.jpg","isAccessibleForFree":true,"articleSection":"Robert Youssef","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":"Robert Youssef","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/cd13aa4e85c346c5904f12ec6310cc23","url":"https://daily.dev/sources/rryssf_"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/google-research-just-proved-you-can-boost-llm-accuracy-by-up-to-76-percentage-points-with-zero-extra-2a4odqkz8","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,google,llm,prompt-engineering"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Robert Youssef","item":"https://daily.dev/sources/rryssf_"},{"@type":"ListItem","position":3,"name":"Google Research just proved you can boost llm accuracy by up to 76 percentage points with zero extra output tokens, zero latency increase, and zero fine-tuning 🤯\n\nthe technique: paste your prompt twice.\n\nthat's it. that's the paper.\n\nbut WHY it works reveals something important about how every llm you use actually reads your input:"}]} ```