{"id":6881,"date":"2025-06-12T05:57:00","date_gmt":"2025-06-11T20:57:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6881"},"modified":"2025-06-07T20:59:14","modified_gmt":"2025-06-07T11:59:14","slug":"pitfalls-in-evaluating-language-model-forecasters","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6881","title":{"rendered":"Pitfalls in Evaluating Language Model Forecasters"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Pitfalls in Evaluating Language Model Forecasters\u00a0<\/strong>[45.4]<br>\u6211\u3005\u306f\u30b3\u30df\u30e5\u30cb\u30c6\u30a3\u3068\u3057\u3066\u3001\u5927\u304d\u306a\u8a00\u8a9e\u30e2\u30c7\u30eb\u3092\u8a55\u4fa1\u3059\u308b\u3088\u3046\u306a\u7d50\u8ad6\u306b\u6ce8\u610f\u3059\u308b\u5fc5\u8981\u304c\u3042\u308b\u3068\u8ad6\u3058\u3066\u3044\u308b\u3002 1) \u6642\u9593\u7684\u30ea\u30fc\u30af\u306b\u3088\u308b\u8a55\u4fa1\u7d50\u679c\u306e\u4fe1\u983c\u306e\u96e3\u3057\u3055,(2) \u8a55\u4fa1\u6027\u80fd\u304b\u3089\u5b9f\u4e16\u754c\u306e\u4e88\u6e2c\u3078\u306e\u5916\u633f\u306e\u96e3\u3057\u3055,\u306e2\u3064\u306e\u30ab\u30c6\u30b4\u30ea\u3092\u8b58\u5225\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2506.00723v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2506.00723v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Sat, 31 May 2025 21:49:17 GMT)<\/li>\n\n\n\n<li>LLM\u306e\u8a55\u4fa1\u306b\u95a2\u3059\u308b\u843d\u3068\u3057\u7a74\u3092\u307e\u3068\u3081\u305f\u8ad6\u6587<\/li>\n\n\n\n<li>\u300cWe identify two broad categories of issues: (1) difficulty in trusting evaluation results due to many forms of temporal leakage, and (2) difficulty in extrapolating from evaluation performance to real-world forecasting. Through systematic analysis and concrete examples from prior work, we demonstrate how evaluation flaws can raise concerns about current and future performance claims.\u300d\u3068\u3044\u3046\u307e\u3068\u3081\u3060\u304c\u3001\u8a55\u4fa1\u306f\u672c\u5f53\u306b\u96e3\u3057\u3044\u3002<\/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":[223,647],"class_list":["post-6881","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-647"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6881","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=6881"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6881\/revisions"}],"predecessor-version":[{"id":6882,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6881\/revisions\/6882"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6881"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6881"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6881"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}