{"id":7130,"date":"2025-07-21T05:48:00","date_gmt":"2025-07-20T20:48:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7130"},"modified":"2025-07-20T14:53:16","modified_gmt":"2025-07-20T05:53:16","slug":"modeling-open-world-cognition-as-on-demand-synthesis-of-probabilistic-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7130","title":{"rendered":"Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models\u00a0<\/strong>[90.7]<br>\u6211\u3005\u306f\u3001\u4eba\u3005\u304c\u5206\u6563\u8868\u73fe\u3068\u8c61\u5fb4\u8868\u73fe\u306e\u7d44\u307f\u5408\u308f\u305b\u3092\u4f7f\u3063\u3066\u3001\u65b0\u3057\u3044\u72b6\u6cc1\u306b\u5408\u308f\u305b\u305f\u898b\u77e5\u3089\u306c\u7cbe\u795e\u30e2\u30c7\u30eb\u3092\u69cb\u7bc9\u3059\u308b\u3068\u3044\u3046\u4eee\u8aac\u3092\u63a2\u6c42\u3059\u308b\u3002 \u30e2\u30c7\u30eb\u5408\u6210\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u300d\u3068\u3044\u3046\u6982\u5ff5\u306e\u8a08\u7b97\u7684\u5b9f\u88c5\u3092\u63d0\u6848\u3059\u308b\u3002 \u6211\u3005\u306f\u3001\u65b0\u3057\u3044\u63a8\u8ad6\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306b\u57fa\u3065\u304f\u4eba\u9593\u306e\u5224\u65ad\u306e\u30e2\u30c7\u30eb\u3068\u3057\u3066\u3001MSA\u3092\u8a55\u4fa1\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2507.12547v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2507.12547v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 16 Jul 2025 18:01:03 GMT)<\/li>\n\n\n\n<li>\u4eba\u9593\u304c\u5185\u5fc3\u7684\u306b\u4f5c\u3063\u3066\u3044\u308b\u30e2\u30c7\u30eb\u3092\u518d\u73fe\u3057\u3088\u3046\u3068\u3059\u308b\u8a66\u307f\u3002<\/li>\n\n\n\n<li>\u300cWe implement an MSA using LMs to parameterize global relevance functions and Probabilistic Programs to construct arbitrary probabilistic models.\u300d\u3068\u3057\u3066\u8a66\u884c\u300cwe find that human reasoning is well-captured by our Model Synthesis Architecture, which provides a better match to human judgments than LM-only baselines and model ablations.\u300d\u3068\u306e\u3053\u3068\u3002\n<ul class=\"wp-block-list\">\n<li>MSA: Model Synthesis Architecture<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[16],"class_list":["post-7130","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-agi"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7130","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=7130"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7130\/revisions"}],"predecessor-version":[{"id":7131,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7130\/revisions\/7131"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7130"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7130"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7130"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}