{"id":6134,"date":"2026-10-19T16:49:15","date_gmt":"2026-10-19T16:49:15","guid":{"rendered":"https:\/\/i10xnews.kinsta.cloud\/?p=6134"},"modified":"2026-10-19T17:19:32","modified_gmt":"2026-10-19T17:19:32","slug":"sharpening-tax-rl-post-training-agentic-coverage","status":"publish","type":"post","link":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage","title":{"rendered":"RL Post-Training Makes Agents More Accurate \u2014 and Less Able to Explore"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">\u26a1 Quick Take<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>The finding:<\/strong> Pre-trained LLMs with a light inference harness often beat their RL post-trained counterparts on solution coverage (pass@K) given enough test-time budget, despite much lower pass@1.<\/li><li><strong>Polarization:<\/strong> Post-training pushes tasks toward always-solved or never-solved. On WebShop with gemma-4-31B, &#8220;pass given compute&#8221; falls from 87.6% to 30.0%, while always-fail rises from 12.4% to 44.0%.<\/li><li><strong>What it means:<\/strong> The authors propose Sharpening Tax to quantify lost test-time scalability after post-training \u2014 prevalent across 42 cases.<\/li><\/ul>\n\n\n<p class=\"wp-block-paragraph\">A popular story about RL post-training is that it mainly sharpens what the base model already knew. Meta Superintelligence Labs, Wisconsin, NYU and Stanford researchers ask whether that story holds for agentic tasks with multi-turn tool use \u2014 and find that base models can still cover more solutions at high sampling budgets.<\/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\/sharpening-tax-rl-post-training-agentic-coverage-webshop.webp\" alt=\"Dumbbell: pass-given-compute falls from 87.6 to 30.0 percent after RL while always-fail rises from 12.4 to 44.0 percent.\" class=\"wp-image-6133\" srcset=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-webshop.webp 1344w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-webshop-300x171.webp 300w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-webshop-1024x585.webp 1024w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-webshop-768x439.webp 768w\" sizes=\"auto, (max-width: 1344px) 100vw, 1344px\" \/><figcaption class=\"wp-element-caption\">i10X chart based on data from Oh et al. (2026), arXiv:2610.01509, Figure 4 \/ Table 6 WebShop analysis (CC BY 4.0).<\/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\">14 base\/post-trained pairs from four families on three agentic benchmarks (42 cases). As budgets grow to 128 rollouts, base pass@K curves often cross and surpass post-trained ones (e.g., over 85% pass@128 on WebShop vs 56% for RL with gemma-4-31B).<\/p>\n\n\n<ul class=\"wp-block-list\"><li><strong>Sampling efficiency vs coverage:<\/strong> RL improves consistency and pass@1 by collapsing the middle.<\/li><li><strong>Incorrect modes amplify too:<\/strong> Always-fail grows alongside always-pass, matching recent work that RL amplifies preferred wrong modes.<\/li><li><strong>Cheap diagnostic:<\/strong> Sharpening Tax can be estimated from a few rollouts.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u2705 Why it matters: pass@1 is not the product<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Products that sample multiple trajectories or search at test time may prefer a base policy&#8217;s broader coverage over an RL policy&#8217;s peak accuracy.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83e\uddea Method in brief<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Pass@1 vs pass@K comparisons; task categorization into always pass \/ pass given compute \/ always fail; Sharpening Tax diagnostic.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2753 What remains open<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Harness dependence:<\/strong> Light inference harnesses matter for base agents.<\/li><li><strong>Benchmark set:<\/strong> Three agentic benches; other domains may differ.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\ud83e\udded Context<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Related RL reasoning work in <a href=\"\/news\/category\/studies\/reasoning\">Reasoning<\/a> and <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 post-training teams<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Report Sharpening Tax:<\/strong> Alongside pass@1.<\/li><li><strong>Keep a base checkpoint:<\/strong> For high-budget search.<\/li><li><strong>Watch always-fail growth:<\/strong> Polarization has a downside.<\/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;Sharpening Tax in Post-Training&#8221; (arXiv:2610.01509v1, submitted October 1, 2026). All numbers are taken from the paper&#8217;s text, tables and figures; i10X did not rerun the experiments. The chart was redrawn by i10X from Figure 4 \/ Table 6 WebShop analysis of the paper, which is published under a CC BY 4.0 license.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2d i10x Perspective<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Sharpening Tax gives a name to the hangover after RL: models become more decisive and less exploratory. For agents that can afford many rollouts, that can be the wrong bargain.<\/p>\n\n","protected":false},"excerpt":{"rendered":"<p>Across 14 base\/post-trained pairs on 3 agentic benches, base models often beat RL successors at pass@K. On WebShop, gemma-4-31B&#8217;s &#8220;pass given compute&#8221; shrinks from 87.6% to 30.0% after RL.<\/p>\n","protected":false},"author":2,"featured_media":6284,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"paper_title":"Sharpening Tax in Post-Training","paper_url":"https:\/\/arxiv.org\/abs\/2610.01509","pdf_url":"https:\/\/arxiv.org\/pdf\/2610.01509v1","doi":"","arxiv_id":"2610.01509","authors":"Changdae Oh, Qi Zeng, Qi Qi, Andrey Zhmoginov, Deren Lei, Yun He, Hoang Phan, Hangoo Kang, Azalia Mirhoseini, Sharon Li","institutions":"Meta Superintelligence Labs; University of Wisconsin\u2013Madison; NYU; Stanford University","paper_date":"2026-10-01","venue":"arXiv (cs.AI), version 1","peer_reviewed":false,"code_url":"","data_url":"","funding_coi":"Work done at Meta (noted for some authors).","figure_license":"Chart drawn by i10X from Figure 4 \/ Table 6 WebShop analysis (CC BY 4.0)","paper_license":"CC BY 4.0","replication":"no","study_models":"14 base\/post-trained pairs from four families incl. gemma-4-31B","study_sample":"3 agentic benchmarks; 42 cases total","footnotes":"","_yoast_wpseo_primary_category":""},"categories":[56,52],"tags":[],"companies":[37,42],"products":[],"class_list":["post-6134","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-reasoning","category-studies","company-google","company-meta"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News<\/title>\n<meta name=\"description\" content=\"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.\" \/>\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\/sharpening-tax-rl-post-training-agentic-coverage\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News\" \/>\n<meta property=\"og:description\" content=\"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage\" \/>\n<meta property=\"og:site_name\" content=\"i10X News\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-19T16:49:15+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-10-19T17:19:32+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-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\\\/sharpening-tax-rl-post-training-agentic-coverage#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\"},\"headline\":\"RL Post-Training Makes Agents More Accurate \u2014 and Less Able to Explore\",\"datePublished\":\"2026-10-19T16:49:15+00:00\",\"dateModified\":\"2026-10-19T17:19:32+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage\"},\"wordCount\":444,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/sharpening-tax-rl-post-training-agentic-coverage-title.webp\",\"articleSection\":[\"Reasoning\",\"Studies\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage\",\"name\":\"Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/sharpening-tax-rl-post-training-agentic-coverage-title.webp\",\"datePublished\":\"2026-10-19T16:49:15+00:00\",\"dateModified\":\"2026-10-19T17:19:32+00:00\",\"description\":\"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/sharpening-tax-rl-post-training-agentic-coverage-title.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/sharpening-tax-rl-post-training-agentic-coverage-title.webp\",\"width\":1344,\"height\":768,\"caption\":\"Illustration: i10X\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/sharpening-tax-rl-post-training-agentic-coverage#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/news.i10x.ai\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"RL Post-Training Makes Agents More Accurate \u2014 and Less Able to Explore\"}]},{\"@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":"Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News","description":"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.","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\/sharpening-tax-rl-post-training-agentic-coverage","og_locale":"en_US","og_type":"article","og_title":"Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News","og_description":"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.","og_url":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage","og_site_name":"i10X News","article_published_time":"2026-10-19T16:49:15+00:00","article_modified_time":"2026-10-19T17:19:32+00:00","og_image":[{"width":1344,"height":768,"url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-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\/sharpening-tax-rl-post-training-agentic-coverage#article","isPartOf":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage"},"author":{"name":"Christopher Ort","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a"},"headline":"RL Post-Training Makes Agents More Accurate \u2014 and Less Able to Explore","datePublished":"2026-10-19T16:49:15+00:00","dateModified":"2026-10-19T17:19:32+00:00","mainEntityOfPage":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage"},"wordCount":444,"commentCount":0,"publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"image":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-title.webp","articleSection":["Reasoning","Studies"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#respond"]}]},{"@type":"WebPage","@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage","url":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage","name":"Sharpening Tax: The Coverage Cost of RL Post-Training - i10X News","isPartOf":{"@id":"https:\/\/news.i10x.ai\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage"},"image":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-title.webp","datePublished":"2026-10-19T16:49:15+00:00","dateModified":"2026-10-19T17:19:32+00:00","description":"Sharpening Tax study: RL post-training boosts pass@1 but cuts pass@K coverage on agentic tasks. Metric across 14 model pairs.","breadcrumb":{"@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#primaryimage","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-title.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/sharpening-tax-rl-post-training-agentic-coverage-title.webp","width":1344,"height":768,"caption":"Illustration: i10X"},{"@type":"BreadcrumbList","@id":"https:\/\/i10x.ai\/news\/sharpening-tax-rl-post-training-agentic-coverage#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/news.i10x.ai\/"},{"@type":"ListItem","position":2,"name":"RL Post-Training Makes Agents More Accurate \u2014 and Less Able to Explore"}]},{"@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\/6134","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=6134"}],"version-history":[{"count":2,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6134\/revisions"}],"predecessor-version":[{"id":6285,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6134\/revisions\/6285"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media\/6284"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media?parent=6134"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/categories?post=6134"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/tags?post=6134"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/companies?post=6134"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/products?post=6134"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}