{"id":6807,"date":"2025-06-03T04:16:00","date_gmt":"2025-06-02T19:16:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6807"},"modified":"2025-05-25T09:30:11","modified_gmt":"2025-05-25T00:30:11","slug":"turnaboutllm-a-deductive-reasoning-benchmark-from-detective-games","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6807","title":{"rendered":"lmgame-Bench: How Good are LLMs at Playing Games? \/ TurnaboutLLM: A Deductive Reasoning Benchmark from Detective Games\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>TurnaboutLLM: A Deductive Reasoning Benchmark from Detective Games&nbsp;<\/strong>[9.2]<br>\u672c\u7a3f\u3067\u306f,Large Language Models(LLM)\u306e\u63a8\u8ad6\u80fd\u529b\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306e\u65b0\u3057\u3044\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3068\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067\u3042\u308bTurnaboutLLM\u3092\u7d39\u4ecb\u3059\u308b\u3002 \u3053\u306e\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u306f\u3001\u9577\u3044\u7269\u8a9e\u306e\u6587\u8108\u306e\u4e2d\u3067\u3001\u8a3c\u8a00\u3068\u8a3c\u62e0\u306e\u9593\u306e\u77db\u76fe\u3092\u8b58\u5225\u3059\u308bLLM\u3092\u51e6\u7406\u3057\u307e\u3059\u3002 \u63d0\u6848\u624b\u6cd5\u306f,12\u7a2e\u985e\u306eLLM\u3092\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u4e0a\u3067\u8a55\u4fa1\u3057,\u5c0e\u51fa\u7684\u63a8\u8ad6\u3092\u5411\u4e0a\u3059\u308b\u305f\u3081\u306e\u4e00\u822c\u7684\u306a\u6226\u7565\u306e\u9650\u754c\u3092\u793a\u5506\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2505.15712v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2505.15712v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Wed, 21 May 2025 16:22:32 GMT)<\/li>\n\n\n\n<li>\u9006\u8ee2\u88c1\u5224\u3084\u30c0\u30f3\u30ac\u30f3\u30ed\u30f3\u30d1\u3092\u4f7f\u3063\u305fLLM\u306e\u6027\u80fd\u8a55\u4fa1\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306e\u63d0\u6848\u3002\u653b\u7565\u30b5\u30a4\u30c8\u306a\u3069\u304cLeak\u306b\u306a\u3063\u3066\u3044\u305d\u3046\u3060\u304c\u3001\u7dcf\u5408\u529b\u304c\u8a66\u3055\u308c\u308b\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u306f\u3042\u308b\u3068\u601d\u3046\u3002LRM\u304c\u512a\u52e2\u306a\u7d50\u679c\uff08\u307e\u3041\u305d\u3046\u3060\u308d\u3046\u3068\u601d\u3046\uff09\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/zharry29\/turnabout_llm\">GitHub &#8211; zharry29\/turnabout_llm<\/a><\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>lmgame-Bench: How Good are LLMs at Playing Games?\u00a0<\/strong>[60.0]<br>\u672c\u7a3f\u3067\u306f,\u73fe\u4ee3\u306e\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb (LLM) \u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306b,\u4eba\u6c17\u30b2\u30fc\u30e0\u3092\u4f7f\u7528\u3059\u308b\u4e0a\u3067\u306e\u5927\u304d\u306a\u8ab2\u984c\u306b\u3064\u3044\u3066\u691c\u8a0e\u3059\u308b\u3002 \u6211\u3005\u306flmgame-Bench\u3092\u5c0e\u5165\u3057\u3001\u30b2\u30fc\u30e0\u3092\u4fe1\u983c\u6027\u8a55\u4fa1\u306b\u5909\u63db\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2505.15146v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2505.15146v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 21 May 2025 06:02:55 GMT)<\/li>\n\n\n\n<li>\u3053\u3061\u3089\u3082\u30b2\u30fc\u30e0\u3092\u7528\u3044\u305f\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u30fb\u8a55\u4fa1\u3002\u300cWe study the major challenges in using popular video games to evaluate modern LLMs and find that directly dropping LLMs into games cannot make an effective evaluation, for three reasons: brittle vision perception, prompt sensitivity, and potential data contamination.\u300d\u3068Leak\u306e\u8ab2\u984c\u304c\u5927\u304d\u3044\u3053\u3068\u3082\u6307\u6458\u3057\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/lmgame-org\/GamingAgent\/\">GitHub &#8211; lmgame-org\/GamingAgent: Computer gaming agents that run on your PC and laptops.<\/a>\u4e0b\u306e<a href=\"https:\/\/github.com\/lmgame-org\/GamingAgent\/lmgame-bench\">https:\/\/github.com\/lmgame-org\/GamingAgent\/lmgame-bench<\/a>\u3068\u306e\u3053\u3068\u3060\u304c\u3001\u73fe\u72b6\u3067\u306f404<\/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":[157,216,517],"class_list":["post-6807","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-game","tag-leakage","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6807","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=6807"}],"version-history":[{"count":2,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6807\/revisions"}],"predecessor-version":[{"id":6812,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6807\/revisions\/6812"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6807"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6807"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6807"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}