{"id":6612,"date":"2025-04-23T04:39:00","date_gmt":"2025-04-22T19:39:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6612"},"modified":"2025-04-23T04:39:00","modified_gmt":"2025-04-22T19:39:00","slug":"xverify-efficient-answer-verifier-for-reasoning-model-evaluations","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6612","title":{"rendered":"xVerify: Efficient Answer Verifier for Reasoning Model Evaluations"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>xVerify: Efficient Answer Verifier for Reasoning Model Evaluations\u00a0<\/strong>[24.1]<br>\u63a8\u8ad6\u30e2\u30c7\u30eb\u8a55\u4fa1\u306e\u305f\u3081\u306e\u52b9\u7387\u7684\u306a\u7b54\u3048\u691c\u8a3c\u5668\u3067\u3042\u308bxVerify\u3092\u63d0\u6848\u3059\u308b\u3002 xVerify\u306f\u540c\u5024\u5224\u5b9a\u306b\u304a\u3044\u3066\u5f37\u3044\u80fd\u529b\u3092\u793a\u3057\u3001\u63a8\u8ad6\u30e2\u30c7\u30eb\u306b\u3088\u3063\u3066\u751f\u6210\u3055\u308c\u305f\u7b54\u3048\u304c\u53c2\u7167\u56de\u7b54\u3068\u7b49\u4fa1\u3067\u3042\u308b\u304b\u3069\u3046\u304b\u3092\u52b9\u679c\u7684\u306b\u6c7a\u5b9a\u3067\u304d\u308b\u3002 \u30c6\u30b9\u30c8\u30bb\u30c3\u30c8\u3068\u4e00\u822c\u5316\u30bb\u30c3\u30c8\u306e\u4e21\u65b9\u3067\u5b9f\u65bd\u3055\u308c\u305f\u8a55\u4fa1\u5b9f\u9a13\u3067\u306f\u3001\u3059\u3079\u3066\u306exVerify\u30e2\u30c7\u30eb\u304c\u5168\u4f53\u306eF1\u30b9\u30b3\u30a2\u306895%\u3092\u8d85\u3048\u308b\u7cbe\u5ea6\u3092\u9054\u6210\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2504.10481v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2504.10481v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 14 Apr 2025 17:59:36 GMT)<\/li>\n\n\n\n<li>LRM\u5411\u3051\u306e\u300cVerify Answer for Reasoning (VAR) dataset\u300d\u3068\u56de\u7b54\u691c\u8a3c\u30e2\u30c7\u30eb\u306e\u63d0\u6848\u3002\u300cxVerify demonstrates strong capability in equivalence judgment, enabling it to effectively determine whether the answers produced by reasoning models are equivalent to reference answers across various types of objective questions.\u300d\u3068\u306e\u3053\u3068\u3067\u3001\u300cxVerify-0.5B-I, outperforms all evaluation methods except GPT-4o, while xVerify-3B-Ib surpasses GPT-4o in overall performance.\u300d\u3068\u3044\u3046\u6027\u80fd\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/IAAR-Shanghai\/xVerify\">GitHub &#8211; IAAR-Shanghai\/xVerify: xVerify: Efficient Answer Verifier for Reasoning Model Evaluations<\/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":[647],"class_list":["post-6612","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-647"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6612","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=6612"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6612\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6612"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6612"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6612"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}