{"id":6154,"date":"2026-10-19T16:49:31","date_gmt":"2026-10-19T16:49:31","guid":{"rendered":"https:\/\/i10xnews.kinsta.cloud\/?p=6154"},"modified":"2026-10-19T17:22:25","modified_gmt":"2026-10-19T17:22:25","slug":"scaling-same-family-on-policy-distillation","status":"publish","type":"post","link":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation","title":{"rendered":"A Small RL Teacher Can Lift a Larger Student Past Itself via On-Policy Distillation"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">\u26a1 Quick Take<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>The finding:<\/strong> In every observed weak-to-strong OPD pair, the student&#8217;s peak held-out gold score exceeds its teacher&#8217;s \u2014 so a compact RL expert can transfer capability upward.<\/li><li><strong>Useful-transfer regime:<\/strong> Early training raises gold score roughly linearly in \u221aKL from the student initialization.<\/li><li><strong>Scale laws:<\/strong> Peak gold improves with teacher scale only up to roughly the student&#8217;s scale; same-base Vanilla-OPD peaks within 0.5 pp of direct-RL references.<\/li><\/ul>\n\n\n<p class=\"wp-block-paragraph\">RL builds strong reasoners, but how that capability transfers across sizes is unclear. Zhejiang University and Kuaishou researchers map on-policy distillation scaling across weak-to-strong, same-base and strong-to-weak setups.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1344\" height=\"768\" src=\"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-peaks.webp\" alt=\"Dumbbell: 7B student peaks from 72.7 to 81.9 percent and 14B from 77.7 to 87.3 as teacher scale grows.\" class=\"wp-image-6153\" srcset=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-peaks.webp 1344w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-peaks-300x171.webp 300w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-peaks-1024x585.webp 1024w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-peaks-768x439.webp 768w\" sizes=\"auto, (max-width: 1344px) 100vw, 1344px\" \/><figcaption class=\"wp-element-caption\">i10X chart based on data from Bao et al. (2026), arXiv:2609.32722, weak-to-strong peak results (arXiv non-exclusive license).<\/figcaption><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2c What the study shows<\/h2>\n\n\n<p class=\"wp-block-paragraph\">They fit power laws for G_peak and the useful-transfer slope versus student\/teacher sizes and teacher gold score. Example: 7B student peaks rise from 72.7% to 81.9% with larger teachers; 14B from 77.7% to 87.3%.<\/p>\n\n\n<ul class=\"wp-block-list\"><li><strong>Off-policy cold starts hurt more as gaps grow:<\/strong> Costs of 6.6\u201315.7+ points appear in weak-to-strong gaps.<\/li><li><strong>Transfer endpoint:<\/strong> Defined when trajectories leave a 95% predictive band for three checkpoints; 9\/25 Vanilla-OPD runs depart within support.<\/li><li><strong>Predictive check:<\/strong> Joint law predicts a held-out peak within 0.3 points (74.1 vs 73.8).<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u2705 Why it matters: distill up, not only down<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Conventional wisdom emphasizes strong-to-weak distillation. This work shows weak-to-strong OPD can overshoot the teacher \u2014 useful when RL is affordable only at small scale.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83e\uddea Method in brief<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Token-level reverse KL diagnostics; power-law fits; Vanilla-OPD vs OffPD comparisons.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2753 What remains open<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Same-family setups:<\/strong> Cross-family transfer untested here.<\/li><li><strong>Gold score definition:<\/strong> Held-out accuracy metric as defined by authors.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\ud83e\udded Context<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Related OPD work in <a href=\"\/news\/category\/studies\/efficiency-training\">Efficiency &amp; Training<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udee0\ufe0f What it means for training stacks<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Try small RL teachers on larger students:<\/strong> They can overshoot.<\/li><li><strong>Watch \u221aKL curves:<\/strong> Useful-transfer ends predictably.<\/li><li><strong>Avoid off-policy cold starts at large gaps:<\/strong> They grow costly.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u270d\ufe0f About the analysis<\/h2>\n\n\n<p class=\"wp-block-paragraph\">This article is based on the full text of the preprint &#8220;Scaling Properties of Same-Family On-Policy Distillation&#8221; (arXiv:2609.32722v1, submitted September 26, 2026). All numbers are taken from the paper&#8217;s text, tables and figures; i10X did not rerun the experiments. The paper is distributed under arXiv&#8217;s non-exclusive license, so i10X does not reproduce its figures; the chart was drawn by i10X from numbers reported in weak-to-strong peak results.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2d i10x Perspective<\/h2>\n\n\n<p class=\"wp-block-paragraph\">The punchline \u2014 every weak-to-strong student beats its teacher \u2014 reframes OPD as an upward capability pump, not just compression. Scaling laws that cap teacher usefulness at student size are the practical constraint to design around.<\/p>\n\n","protected":false},"excerpt":{"rendered":"<p>ZJU\/Kuaishou study: in every weak-to-strong OPD pair, the student&#8217;s peak gold score exceeds the teacher&#8217;s. Peak scores rise with teacher scale only up to about the student&#8217;s size.<\/p>\n","protected":false},"author":2,"featured_media":6304,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"paper_title":"Scaling Properties of Same-Family On-Policy Distillation","paper_url":"https:\/\/arxiv.org\/abs\/2609.32722","pdf_url":"https:\/\/arxiv.org\/pdf\/2609.32722v1","doi":"","arxiv_id":"2609.32722","authors":"Yuntai Bao, Qinfeng Li, Guoqing Jiang, Liwei Chen, Zhiheng Qin, Xuanping Li, Wenqi Zhang, Xuhong Zhang","institutions":"Zhejiang University; Kuaishou Technology","paper_date":"2026-09-26","venue":"arXiv (cs.LG), version 1","peer_reviewed":false,"code_url":"","data_url":"","funding_coi":"Internship at Kuaishou Technology.","figure_license":"Chart drawn by i10X from weak-to-strong peak results (arXiv non-exclusive license)","paper_license":"arXiv non-exclusive license","replication":"no","study_models":"Same-family teacher\u2013student pairs across multiple scales (0.5B\u201314B class)","study_sample":"Weak-to-strong, same-base and strong-to-weak OPD runs with held-out gold scores","footnotes":"","_yoast_wpseo_primary_category":""},"categories":[57,52],"tags":[],"companies":[],"products":[],"class_list":["post-6154","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-efficiency-training","category-studies"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Scaling Laws for Same-Family On-Policy Distillation - i10X News<\/title>\n<meta name=\"description\" content=\"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Scaling Laws for Same-Family On-Policy Distillation - i10X News\" \/>\n<meta property=\"og:description\" content=\"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation\" \/>\n<meta property=\"og:site_name\" content=\"i10X News\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-19T16:49:31+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-10-19T17:22:25+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1344\" \/>\n\t<meta property=\"og:image:height\" content=\"768\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Christopher Ort\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@i10x_ai\" \/>\n<meta name=\"twitter:site\" content=\"@i10x_ai\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Christopher Ort\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\"},\"headline\":\"A Small RL Teacher Can Lift a Larger Student Past Itself via On-Policy Distillation\",\"datePublished\":\"2026-10-19T16:49:31+00:00\",\"dateModified\":\"2026-10-19T17:22:25+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation\"},\"wordCount\":421,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/scaling-same-family-on-policy-distillation-title.webp\",\"articleSection\":[\"Efficiency &amp; Training\",\"Studies\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation\",\"name\":\"Scaling Laws for Same-Family On-Policy Distillation - i10X News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/scaling-same-family-on-policy-distillation-title.webp\",\"datePublished\":\"2026-10-19T16:49:31+00:00\",\"dateModified\":\"2026-10-19T17:22:25+00:00\",\"description\":\"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/scaling-same-family-on-policy-distillation-title.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/scaling-same-family-on-policy-distillation-title.webp\",\"width\":1344,\"height\":768,\"caption\":\"Illustration: i10X\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/scaling-same-family-on-policy-distillation#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/news.i10x.ai\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"A Small RL Teacher Can Lift a Larger Student Past Itself via On-Policy Distillation\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\",\"url\":\"https:\\\/\\\/news.i10x.ai\\\/\",\"name\":\"i10X News\",\"description\":\"Stay updated with the latest news and announcements from i10X.\",\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/news.i10x.ai\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\",\"name\":\"i10X\",\"url\":\"https:\\\/\\\/news.i10x.ai\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/i10x-logo-180.png\",\"contentUrl\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/i10x-logo-180.png\",\"width\":180,\"height\":180,\"caption\":\"i10X\"},\"image\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/i10x_ai\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/i10x-ai\",\"https:\\\/\\\/instagram.com\\\/i10x.ai\",\"https:\\\/\\\/www.youtube.com\\\/@i10X_ai_official\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\",\"name\":\"Christopher Ort\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"caption\":\"Christopher Ort\"},\"description\":\"Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU). He is the author of \u201cThe Style Penalty\u201d, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/author\\\/christopher-ort\",\"sameAs\":[\"https:\\\/\\\/i10x.ai\\\/blog\\\/author\\\/christopher-ort\"],\"worksFor\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#organization\"},\"alumniOf\":{\"@type\":\"EducationalOrganization\",\"name\":\"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU)\",\"url\":\"https:\\\/\\\/www.fau.eu\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Scaling Laws for Same-Family On-Policy Distillation - i10X News","description":"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation","og_locale":"en_US","og_type":"article","og_title":"Scaling Laws for Same-Family On-Policy Distillation - i10X News","og_description":"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.","og_url":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation","og_site_name":"i10X News","article_published_time":"2026-10-19T16:49:31+00:00","article_modified_time":"2026-10-19T17:22:25+00:00","og_image":[{"width":1344,"height":768,"url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp","type":"image\/webp"}],"author":"Christopher Ort","twitter_card":"summary_large_image","twitter_creator":"@i10x_ai","twitter_site":"@i10x_ai","twitter_misc":{"Written by":"Christopher Ort","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#article","isPartOf":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation"},"author":{"name":"Christopher Ort","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a"},"headline":"A Small RL Teacher Can Lift a Larger Student Past Itself via On-Policy Distillation","datePublished":"2026-10-19T16:49:31+00:00","dateModified":"2026-10-19T17:22:25+00:00","mainEntityOfPage":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation"},"wordCount":421,"commentCount":0,"publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"image":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp","articleSection":["Efficiency &amp; Training","Studies"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#respond"]}]},{"@type":"WebPage","@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation","url":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation","name":"Scaling Laws for Same-Family On-Policy Distillation - i10X News","isPartOf":{"@id":"https:\/\/news.i10x.ai\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#primaryimage"},"image":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp","datePublished":"2026-10-19T16:49:31+00:00","dateModified":"2026-10-19T17:22:25+00:00","description":"OPD scaling study: students can beat smaller RL teachers; useful-transfer is linear in sqrt(KL). Power laws for peak gold score.","breadcrumb":{"@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#primaryimage","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/scaling-same-family-on-policy-distillation-title.webp","width":1344,"height":768,"caption":"Illustration: i10X"},{"@type":"BreadcrumbList","@id":"https:\/\/i10x.ai\/news\/scaling-same-family-on-policy-distillation#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/news.i10x.ai\/"},{"@type":"ListItem","position":2,"name":"A Small RL Teacher Can Lift a Larger Student Past Itself via On-Policy Distillation"}]},{"@type":"WebSite","@id":"https:\/\/news.i10x.ai\/#website","url":"https:\/\/news.i10x.ai\/","name":"i10X News","description":"Stay updated with the latest news and announcements from i10X.","publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/news.i10x.ai\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/news.i10x.ai\/#organization","name":"i10X","url":"https:\/\/news.i10x.ai\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/news.i10x.ai\/#\/schema\/logo\/image\/","url":"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/i10x-logo-180.png","contentUrl":"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/i10x-logo-180.png","width":180,"height":180,"caption":"i10X"},"image":{"@id":"https:\/\/news.i10x.ai\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/i10x_ai","https:\/\/www.linkedin.com\/company\/i10x-ai","https:\/\/instagram.com\/i10x.ai","https:\/\/www.youtube.com\/@i10X_ai_official"]},{"@type":"Person","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a","name":"Christopher Ort","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","caption":"Christopher Ort"},"description":"Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU). He is the author of \u201cThe Style Penalty\u201d, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.","url":"https:\/\/i10x.ai\/news\/author\/christopher-ort","sameAs":["https:\/\/i10x.ai\/blog\/author\/christopher-ort"],"worksFor":{"@id":"https:\/\/i10x.ai\/news\/#organization"},"alumniOf":{"@type":"EducationalOrganization","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU)","url":"https:\/\/www.fau.eu"}}]}},"_links":{"self":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6154","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/comments?post=6154"}],"version-history":[{"count":2,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6154\/revisions"}],"predecessor-version":[{"id":6305,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6154\/revisions\/6305"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media\/6304"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media?parent=6154"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/categories?post=6154"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/tags?post=6154"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/companies?post=6154"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/products?post=6154"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}