{"id":6589,"date":"2025-04-22T04:33:00","date_gmt":"2025-04-21T19:33:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6589"},"modified":"2025-04-22T04:33:00","modified_gmt":"2025-04-21T19:33:00","slug":"crosswordbench-evaluating-the-reasoning-capabilities-of-llms-and-lvlms-with-controllable-puzzle-generation","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6589","title":{"rendered":"CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation\u00a0<\/strong>[53.5]<br>CrossWordBench\u306f\u3001\u5927\u304d\u306a\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u3068LVLM(Large Vision-Language Models)\u306e\u63a8\u8ad6\u80fd\u529b\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306b\u8a2d\u8a08\u3055\u308c\u305f\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308b\u3002 \u8a55\u4fa1\u306e\u7d50\u679c,LLM\u306e\u63a8\u8ad6\u306f,\u30af\u30ed\u30b9\u30ec\u30bf\u30fc\u5236\u7d04\u3092\u52b9\u679c\u7684\u306b\u6d3b\u7528\u3059\u308b\u3053\u3068\u306b\u3088\u308a,\u975e\u63a8\u8ad6\u30e2\u30c7\u30eb\u3088\u308a\u3082\u5927\u5e45\u306b\u512a\u308c\u3066\u3044\u308b\u3053\u3068\u304c\u308f\u304b\u3063\u305f\u3002 \u672c\u7814\u7a76\u306f,\u73fe\u5728\u306eLLM\u3068LVLM\u306e\u63a8\u8ad6\u80fd\u529b\u306e\u9650\u754c\u306b\u3064\u3044\u3066\u8003\u5bdf\u3057,\u4eca\u5f8c\u306e\u8a55\u4fa1\u306e\u305f\u3081\u306b,\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u5236\u7d04\u30bf\u30b9\u30af\u3092\u4f5c\u6210\u3059\u308b\u305f\u3081\u306e\u52b9\u679c\u7684\u306a\u30a2\u30d7\u30ed\u30fc\u30c1\u3092\u63d0\u4f9b\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2504.00043v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2504.00043v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Sun, 30 Mar 2025 20:03:36 GMT)<\/li>\n\n\n\n<li>\u30af\u30ed\u30b9\u30ef\u30fc\u30c9\u30d1\u30ba\u30eb\u3092\u7528\u3044\u308b\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u300cCrossWordBench collects data and generates puzzles from three sources: (1) multilingual word-clue pairs from public repositories, (2) dictionary-based definitions, and (3) adapted questions-answer pairs from existing benchmarks (e g , CommonsenseQA (Talmor et al , 2018)) where the answers are open-ended or unconstrained.\u300d\u3068\u3044\u3046\u69cb\u7bc9\u65b9\u91dd\u3002\u7d50\u679c\u306f\u300cOur extensive evaluation of over 20 models shows that reasoning models substantially outperform non-reasoning counterparts and can benefit from increased crossing-letter constraints.\u300d\u3068LRM\u306f\u5f37\u3044<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/SeanLeng1\/CrossWordBench\">GitHub &#8211; SeanLeng1\/CrossWordBench<\/a>\u3001<a href=\"https:\/\/huggingface.co\/datasets\/HINT-lab\/CrossWordBench\">HINT-lab\/CrossWordBench \u00b7 Datasets at Hugging Face<\/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":[223,434,517],"class_list":["post-6589","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-vision-language","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6589","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=6589"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6589\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6589"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6589"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6589"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}