{"id":4303,"date":"2024-01-15T05:55:00","date_gmt":"2024-01-14T20:55:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4303"},"modified":"2024-01-15T05:55:00","modified_gmt":"2024-01-14T20:55:00","slug":"risk-taxonomy-mitigation-and-assessment-benchmarks-of-large-language-model-systems","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4303","title":{"rendered":"TrustLLM\u3068LLM\u306e\u30ea\u30b9\u30af\u5206\u985e"},"content":{"rendered":"\n<p>LLM\u306e\u4fe1\u983c\u6027\u3001\u5b89\u5168\u6027\u306b\u95a2\u3059\u308b\u8ad6\u6587\u3002TrustLLM\u306f\u8457\u8005\u6570\u304c\u3059\u3054\u3044\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>TrustLLM: Trustworthiness in Large Language Models\u00a0<\/strong>[446.2]<br>\u672c\u7a3f\u3067\u306f,\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u306b\u304a\u3051\u308b\u4fe1\u983c\u5ea6\u306b\u95a2\u3059\u308b\u7dcf\u5408\u7684\u7814\u7a76\u3067\u3042\u308bTrustLLM\u3092\u7d39\u4ecb\u3059\u308b\u3002 \u307e\u305a\u30018\u3064\u306e\u7570\u306a\u308b\u6b21\u5143\u306b\u307e\u305f\u304c\u308b\u4fe1\u983c\u6027\u306e\u9ad8\u3044LCM\u306e\u539f\u5247\u3092\u63d0\u6848\u3059\u308b\u3002 \u3053\u308c\u3089\u306e\u539f\u5247\u306b\u57fa\u3065\u3044\u3066\u3001\u771f\u7406\u6027\u3001\u5b89\u5168\u6027\u3001\u516c\u6b63\u6027\u3001\u5805\u7262\u6027\u3001\u30d7\u30e9\u30a4\u30d0\u30b7\u3001\u6a5f\u68b0\u502b\u7406\u3092\u542b\u30806\u3064\u306e\u6b21\u5143\u306b\u308f\u305f\u308b\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3092\u78ba\u7acb\u3057\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2401.05561v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2401.05561v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 10 Jan 2024 22:07:21 GMT)<\/li>\n\n\n\n<li>LLM\u306e\u4fe1\u983c\u6027\u306b\u95a2\u3059\u308b\u5305\u62ec\u7684\u306a\u30b5\u30fc\u30d9\u30a4<\/li>\n\n\n\n<li>\u300c\u00a0\u201cto be trustworthy, LLMs must appropriately reflect characteristics such as truthfulness, safety, fairness, robustness, privacy, machine ethics, transparency, and accountability.\u201d\u300d\u3092\u30b9\u30bf\u30fc\u30c8\u306b\u3057\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li>\u30d7\u30ed\u30b8\u30a7\u30af\u30c8\u30b5\u30a4\u30c8\u306f<a href=\"https:\/\/trustllmbenchmark.github.io\/TrustLLM-Website\/\">TrustLLM-Benchmark (trustllmbenchmark.github.io)<\/a><\/li>\n<\/ul>\n\n\n\n<p><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems\u00a0<\/strong>[29.8]<br>\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u306f\u3001\u591a\u69d8\u306a\u81ea\u7136\u8a00\u8a9e\u51e6\u7406\u30bf\u30b9\u30af\u3092\u89e3\u304f\u4e0a\u3067\u5f37\u529b\u306a\u80fd\u529b\u3092\u6301\u3064\u3002 \u3057\u304b\u3057\u3001LLM\u30b7\u30b9\u30c6\u30e0\u306e\u5b89\u5168\u6027\u3068\u30bb\u30ad\u30e5\u30ea\u30c6\u30a3\u306e\u554f\u984c\u306f\u3001\u305d\u306e\u5e83\u7bc4\u306a\u5fdc\u7528\u306b\u3068\u3063\u3066\u5927\u304d\u306a\u969c\u5bb3\u3068\u306a\u3063\u3066\u3044\u308b\u3002 \u672c\u7a3f\u3067\u306f,LLM\u30b7\u30b9\u30c6\u30e0\u306e\u5404\u30e2\u30b8\u30e5\u30fc\u30eb\u306b\u95a2\u9023\u3059\u308b\u6f5c\u5728\u7684\u306a\u30ea\u30b9\u30af\u3092\u4f53\u7cfb\u7684\u306b\u5206\u6790\u3059\u308b\u5305\u62ec\u7684\u5206\u985e\u6cd5\u3092\u63d0\u6848\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2401.05778v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2401.05778v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 11 Jan 2024 09:29:56 GMT)<\/li>\n\n\n\n<li>LLM\u306e\u30ea\u30b9\u30af\u306b\u95a2\u3059\u308b\u5206\u6790\u3068\u5206\u985e\u3001\u300c\u5165\u529b\u30e2\u30b8\u30e5\u30fc\u30eb\u300d\u300c\u8a00\u8a9e\u30e2\u30c7\u30eb\u300d\u300c\u30c4\u30fc\u30eb\u30c1\u30a7\u30a4\u30f3\u300d\u300c\u51fa\u529b\u30e2\u30b8\u30e5\u30fc\u30eb\u300d\u306e\uff14\u3064\u3092\u5bfe\u8c61\u3068\u3057\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li>\u975e\u5e38\u306b\u3088\u304f\u307e\u3068\u307e\u3063\u3066\u3044\u308b\u306e\u3068\u30e9\u30a4\u30bb\u30f3\u30b9\u304cCC-BY\u3068\u3044\u3046\u306e\u304c\u3042\u308a\u304c\u305f\u3044\u3002<\/li>\n<\/ul>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>LLM\u306e\u4fe1\u983c\u6027\u3001\u5b89\u5168\u6027\u306b\u95a2\u3059\u308b\u8ad6\u6587\u3002TrustLLM\u306f\u8457\u8005\u6570\u304c\u3059\u3054\u3044\u3002<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[223,387,622],"class_list":["post-4303","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-survey","tag-622"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4303","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=4303"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4303\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4303"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4303"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4303"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}