
{"id":11140,"date":"2026-08-18T02:50:26","date_gmt":"2026-08-17T18:50:26","guid":{"rendered":"https:\/\/infernews.com\/blog\/littlelearner\/"},"modified":"2026-08-18T02:50:26","modified_gmt":"2026-08-17T18:50:26","slug":"littlelearner","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/littlelearner\/","title":{"rendered":"LittleLearner \u628a\u8a9e\u8a00\u6a21\u578b\u7684\u77e5\u8b58\u908a\u754c\u63a7\u5236\u5728\u5c0f\u5b78\u4e94\u5e74\u7d1a"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/08\/pasted-4743c7309da3.jpg\" alt=\"Hero image preview\"><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">\u6a21\u578b\u80fd\u5426\u56de\u7b54\u554f\u984c\uff0c\u4e0d\u4e00\u5b9a\u4ee3\u8868\u5b83\u771f\u6b63\u5b78\u904e\u76f8\u95dc\u77e5\u8b58\u3002LittleLearner \u5c07\u8a9e\u8a00\u6a21\u578b\uff08Language Models\uff0cLMs\uff09\u7684\u9810\u8a13\u7df4\u8cc7\u6599\u9650\u5236\u5728\u7f8e\u570b\u5c0f\u5b78 K\u20135 \u8ab2\u7a0b\uff0c\u5efa\u7acb\u4e00\u500b\u53ef\u89c0\u5bdf\u77e5\u8b58\u908a\u754c\u7684\u7814\u7a76\u6c99\u76d2\uff0c\u8b93\u7814\u7a76\u8005\u5206\u8fa8\u80fd\u529b\u662f\u5f9e\u8cc7\u6599\u4e2d\u7372\u5f97\uff0c\u9084\u662f\u53ea\u662f\u5728\u63d0\u793a\u4e0b\u88ab\u5f15\u51fa\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u9805\u76ee\u7684\u6838\u5fc3\u662f LittleCurriculum\uff0c\u4e00\u500b\u7531 FineWeb-Edu \u84b8\u993e\u800c\u6210\u3001\u7d04 880 \u5104\u500b token \u7684\u8a9e\u6599\u5eab\u3002\u8cc7\u6599\u7d93\u904e\u4e94\u968e\u6bb5\u7be9\u9078\uff0c\u4e26\u6309\u7167 Common Core standards \u5c0d\u9f4a\uff0c\u660e\u78ba\u6392\u9664\u4e94\u5e74\u7d1a\u4ee5\u4e0a\u6559\u6388\u7684\u6982\u5ff5\u3001\u4e8b\u5be6\u548c\u8a5e\u5f59\uff1b\u6a21\u578b\u5247\u5f9e\u96f6\u958b\u59cb\u8a13\u7df4\uff0c\u4e26\u914d\u5099\u4f7f\u7528\u76f8\u540c\u67b6\u69cb\u3001token \u6578\u91cf\u548c\u8a13\u7df4\u914d\u65b9\u7684 Unfiltered control\uff0c\u65b9\u4fbf\u4f5c\u51fa\u4e7e\u6de8\u6bd4\u8f03\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">LittleLearner \u63d0\u4f9b 0.6B\u30011.3B \u548c 5B \u4e09\u7a2e\u898f\u6a21\uff0c\u53e6\u6709 Base\u3001GRPO \u53ca Chatty \u7248\u672c\u3002\u7814\u7a76\u6e2c\u8a66 scaling\u3001SFT+GRPO post-training \u548c in-context learning \u5f8c\u767c\u73fe\uff0c\u9019\u4e9b\u65b9\u6cd5\u53ef\u4ee5\u653e\u5927\u8ab2\u7a0b\u7bc4\u570d\u5167\u7684\u80fd\u529b\uff0c\u537b\u672a\u80fd\u660e\u986f\u63d0\u5347\u7bc4\u570d\u4ee5\u5916\u7684\u8868\u73fe\uff0c\u986f\u793a\u9810\u8a13\u7df4\u8cc7\u6599\u53ef\u80fd\u8a2d\u5b9a\u4e86\u6709\u6548\u80fd\u529b\u4e0a\u9650\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u91cd\u9ede\u5305\u62ec\uff1a<br \/>\n&#8211; 88B-token LittleCurriculum \u5c07\u77e5\u8b58\u66b4\u9732\u7bc4\u570d\u63a7\u5236\u5728 K\u20135 \u8ab2\u7a0b<br \/>\n&#8211; \u4e09\u7a2e\u6a21\u578b\u898f\u6a21\u5747\u8a2d\u6709\u5339\u914d\u7684 Unfiltered control<br \/>\n&#8211; scaling \u53ef\u6539\u5584\u8ab2\u7a0b\u7bc4\u570d\u5167\u8868\u73fe\uff0c\u5916\u63a8\u80fd\u529b\u53ea\u6709\u9650\u5ea6\u589e\u52a0<br \/>\n&#8211; SFT+GRPO \u548c in-context learning \u672a\u80fd\u7a81\u7834\u77e5\u8b58\u908a\u754c<br \/>\n&#8211; \u7db2\u9801\u63d0\u4f9b 5B \u6a21\u578b\u7684\u700f\u89bd\u5668\u804a\u5929\u4ecb\u9762\u3001Dataset \u548c Model checkpoints<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u5c0d\u7814\u7a76\u6a21\u578b\u80fd\u529b\u4f86\u6e90\u3001\u77e5\u8b58\u9077\u79fb\u548c post-training \u6548\u679c\u7684\u4eba\uff0cLittleLearner \u63d0\u4f9b\u4e86\u8f03\u5bb9\u6613\u91cd\u73fe\u548c\u6bd4\u8f03\u7684\u5be6\u9a57\u57fa\u790e\uff1b\u5c0d\u4e00\u822c\u4f7f\u7528\u8005\u800c\u8a00\uff0c\u700f\u89bd\u5668\u5167\u7684 live chat \u5247\u53ef\u76f4\u63a5\u611f\u53d7\u4e00\u500b\u53d7\u63a7\u77e5\u8b58\u7bc4\u570d\u6a21\u578b\u7684\u56de\u7b54\u65b9\u5f0f\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/littlelearner-ll.github.io\/\" rel=\"noopener noreferrer\"><strong>\u9805\u76ee\u4e3b\u9801<\/strong><\/a> \u00b7 <a href=\"https:\/\/huggingface.co\/littlelearner\/littlelearner-5b-grpo-math-expert\" rel=\"noopener noreferrer\"><strong>\u6a21\u578b<\/strong><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>LittleLearner \u900f\u904e\u53d7\u63a7\u8ab2\u7a0b\u8cc7\u6599\u8a13\u7df4\u6a21\u578b\uff0c\u7814\u7a76\u80fd\u529b\u7a76\u7adf\u4f86\u81ea\u5b78\u7fd2\uff0c\u9084\u662f\u53ea\u662f\u88ab\u63d0\u793a\u5f15\u51fa\u3002<\/p>\n","protected":false},"author":8,"featured_media":11139,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ai_generated_summary":"","footnotes":""},"categories":[133,76,127,199,201],"tags":[],"class_list":["post-11140","post","type-post","status-publish","format-standard","hentry","category-133","category-76","category-127","category-dataset-","category-skill-"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/11140","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=11140"}],"version-history":[{"count":0,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/11140\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/11139"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=11140"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=11140"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=11140"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}