{"id":3963,"date":"2023-10-23T06:34:00","date_gmt":"2023-10-22T21:34:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=3963"},"modified":"2023-10-23T06:34:00","modified_gmt":"2023-10-22T21:34:00","slug":"set-of-mark-prompting","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=3963","title":{"rendered":"Set-of-Mark Prompting"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Set-of-Mark Prompting Unleashes Extraordinary Visual Grounding in GPT-4V&nbsp;<\/strong>[103.7]<br>\u5927\u898f\u6a21\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u30e2\u30c7\u30eb\u306e\u8996\u899a\u7684\u30b0\u30e9\u30a6\u30f3\u30c9\u5316\u80fd\u529b\u3092\u89e3\u304d\u653e\u3064\u305f\u3081\u306b,\u65b0\u3057\u3044\u8996\u899a\u7684\u30d7\u30ed\u30f3\u30d7\u30c8\u624b\u6cd5\u3067\u3042\u308bSet-of-Mark(SoM)\u3092\u63d0\u6848\u3059\u308b\u3002 \u6211\u3005\u306f\u3001SAM\u306e\u3088\u3046\u306a\u5e02\u8ca9\u306e\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u30bb\u30b0\u30e1\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3\u30e2\u30c7\u30eb\u3092\u7528\u3044\u3066\u3001\u753b\u50cf\u3092\u7570\u306a\u308b\u30ec\u30d9\u30eb\u306e\u7c92\u5ea6\u306e\u9818\u57df\u306b\u5206\u5272\u3057\u3001\u3053\u308c\u3089\u306e\u9818\u57df\u3092\u4e00\u9023\u306e\u30de\u30fc\u30af\u3067\u30aa\u30fc\u30d0\u30fc\u30ec\u30a4\u3059\u308b\u3002 \u30de\u30fc\u30af\u3055\u308c\u305f\u30a4\u30e1\u30fc\u30b8\u3092\u5165\u529b\u3068\u3057\u3066\u4f7f\u7528\u3059\u308b\u3053\u3068\u3067\u3001GPT-4V\u306f\u8996\u899a\u7684\u306a\u63a5\u5730\u3092\u5fc5\u8981\u3068\u3059\u308b\u8cea\u554f\u306b\u7b54\u3048\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2310.11441v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2310.11441v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Tue, 17 Oct 2023 17:51:31 GMT)<\/li>\n\n\n\n<li>GPT-4V\u306b\u5bfe\u3059\u308b\u30d7\u30ed\u30f3\u30d7\u30c8\u30c6\u30af\u30cb\u30c3\u30af\u3001Set-of-Mark\u306e\u63d0\u6848\u3002\u901f\u5ea6\u52dd\u8ca0\u307f\u305f\u3044\u306a\u3068\u3053\u308d\u3082\u3042\u308b\u306e\u3060\u308d\u3046\u3051\u3069\u3001\u8ad6\u6587\u51fa\u308b\u306e\u901f\u3059\u304e\u30fb\u30fb\u30fb<\/li>\n\n\n\n<li>\u300cWe show that simply overlaying a number of symbolic marks on a set of regions of an input image can unleash the visual grounding ability of GPT-4V.\u300d\u3068\u306e\u3053\u3068\u3002\u4eba\u9593\u3067\u3082\u753b\u50cf\u306b\u30ac\u30a4\u30c9\u3092\u5165\u308c\u308b\u3068\u30bf\u30b9\u30af\u3092\u3084\u308a\u3084\u3059\u304f\u306a\u308b\u306e\u3067\u30a2\u30a4\u30c7\u30a2\u3068\u3057\u3066\u306f\u305d\u3046\u3060\u308d\u3046\u3068\u601d\u3046\u3082\u306e\u306e\u3001\u5e83\u7bc4\u306a\u5b9f\u9a13\u30fb\u691c\u8a3c\u306f\u3068\u3066\u3082\u53c2\u8003\u306b\u306a\u308b\u3002<\/li>\n\n\n\n<li>\u30d7\u30ed\u30b8\u30a7\u30af\u30c8\u30b5\u30a4\u30c8\u306f<a href=\"https:\/\/som-gpt4v.github.io\/\">SoM-GPT4V<\/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":[176,314],"class_list":["post-3963","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-gpt-4v","tag-prompting"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3963","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=3963"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3963\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3963"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3963"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3963"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}