{"id":4675,"date":"2024-04-19T05:20:00","date_gmt":"2024-04-18T20:20:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4675"},"modified":"2024-04-19T05:20:00","modified_gmt":"2024-04-18T20:20:00","slug":"jailbreakv-28k","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4675","title":{"rendered":"JailBreakV-28K"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks\u00a0<\/strong>[24.7]<br>\u672c\u7a3f\u3067\u306f,\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u3092\u6210\u529f\u3055\u305b\u308b\u624b\u6cd5\u304c,MLLM\u306e\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30fc\u30af\u306b\u7b49\u3057\u304f\u6709\u52b9\u304b\u3069\u3046\u304b\u3092\u691c\u8a0e\u3059\u308b\u3002 MLLM \u3078\u306e LLM \u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u624b\u6cd5\u306e\u8ee2\u9001\u6027\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306e\u5148\u99c6\u7684\u306a\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308b JailBreakV-28K \u3092\u7d39\u4ecb\u3059\u308b\u3002 LLM\u306e\u9ad8\u5ea6\u306a\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u653b\u6483\u3068\u3001\u6700\u8fd1\u306eMLLM\u306e\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u653b\u6483\u306b\u3088\u308b\u30a4\u30e1\u30fc\u30b8\u30d9\u30fc\u30b9\u306e\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u5165\u529b\u306b\u3088\u308a\u300120000\u306e\u30c6\u30ad\u30b9\u30c8\u30d9\u30fc\u30b9\u306e\u30b8\u30a7\u30a4\u30eb\u30d6\u30ec\u30a4\u30af\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u751f\u6210\u3057\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2404.03027v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2404.03027v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 03 Apr 2024 19:23:18 GMT)<\/li>\n\n\n\n<li>MLLM\u3078\u306eJailbreak\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3002\u300cOur extensive experiments reveal that MLLMs inherit vulnerability from their LLM counterparts.\u300d\u306f\u307e\u3041\u305d\u3046\u3060\u308d\u3046\u3068\u601d\u3044\u3064\u3064\u30fb\u30fb\u30fb\u300cIn addition, text-based jailbreak attacks are more effective than image-based jailbreak attacks and are effective regardless of the image input.\u300d\u306f\u30fb\u30fb\u30fb<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/huggingface.co\/datasets\/JailbreakV-28K\/JailBreakV-28k\">JailbreakV-28K\/JailBreakV-28k \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":[207,251,517],"class_list":["post-4675","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-jailbreak","tag-mllm","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4675","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=4675"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4675\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4675"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4675"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4675"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}