{"id":5987,"date":"2024-12-30T04:28:00","date_gmt":"2024-12-29T19:28:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5987"},"modified":"2024-12-30T04:28:00","modified_gmt":"2024-12-29T19:28:00","slug":"b-star-monitoring-and-balancing-exploration-and-exploitation-in-self-taught-reasoners","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5987","title":{"rendered":"B-STaR: Monitoring and Balancing Exploration and Exploitation in Self-Taught Reasoners\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>B-STaR: Monitoring and Balancing Exploration and Exploitation in Self-Taught Reasoners\u00a0<\/strong>[19.0]<br>\u81ea\u5df1\u6539\u5584\u306f\u3001\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3092\u5411\u4e0a\u3055\u305b\u308b\u4e3b\u8981\u306a\u65b9\u6cd5\u3068\u3057\u3066\u73fe\u308c\u3066\u3044\u307e\u3059\u3002 \u672c\u7a3f\u3067\u306f,\u3053\u306e\u53cd\u5fa9\u7684\u30d7\u30ed\u30bb\u30b9\u306b\u304a\u3044\u30662\u3064\u306e\u91cd\u8981\u306a\u8981\u56e0\u3092\u30e2\u30cb\u30bf\u3059\u308b\u624b\u6cd5\u3092\u63d0\u6848\u3057,\u63d0\u6848\u3059\u308b\u3002 B-STaR\u306f\u3001\u53cd\u5fa9\u7684\u306a\u69cb\u6210\u3092\u8abf\u6574\u3057\u3001\u63a2\u7d22\u3068\u30a8\u30af\u30b9\u30d7\u30ed\u30a4\u30c8\u306e\u30d0\u30e9\u30f3\u30b9\u3092\u3068\u308b\u81ea\u5df1\u5b66\u7fd2\u63a8\u8ad6\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2412.17256v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2412.17256v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 23 Dec 2024 03:58:34 GMT)<\/li>\n\n\n\n<li>\u300cIn this work, we identify and propose methods to monitor two pivotal factors in this iterative process: (1) the model\u2019s ability to generate sufficiently diverse responses (exploration); and (2) the effectiveness of external rewards in distinguishing high-quality candidates from lower-quality ones (exploitation).\u300d\u3001\u306b\u3064\u3044\u3066\u3053\u308c\u3089\u3092\u76e3\u8996\u3057\u30d0\u30e9\u30f3\u30b9\u3092\u3068\u308b\u624b\u6cd5\u3092\u63d0\u6848\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/hkust-nlp\/B-STaR\">GitHub &#8211; hkust-nlp\/B-STaR<\/a><\/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":[356],"class_list":["post-5987","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-self-x"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5987","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=5987"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5987\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5987"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5987"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5987"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}