[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f5RBRsI9JgHjof0Ja8Z3d1IE1mhl40tzGnou-urUneT8":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":16,"related":17,"source":27,"type":28},[],"2025-11-15 11:48:20",1081866465,[8,9,10,11],"仅输入输出行为","模型内部参数和训练数据","模型训练日志","用户隐私数据",{"courseId":13,"workId":14,"workName":15},"1000140560","63213907","第九章单元测试","白盒检测场景下,检测者可以访问以下哪项信息?( )",[18,29,32,40,49],{"answer":19,"createTime":5,"id":20,"options":21,"question":26,"source":27,"type":28},[],1081866458,[22,23,24,25],"提升模型整体准确率","激活预设的异常输出","删除训练数据中的噪声","优化模型训练速度","后门攻击中,触发器的功能是?( )","v2",0,{"answer":30,"createTime":5,"id":6,"options":31,"question":16,"source":27,"type":28},[],[8,9,10,11],{"answer":33,"createTime":5,"id":34,"options":35,"question":38,"source":27,"type":39},[],1081866506,[36,37],"对","错","后门攻击会导致模型对所有输入样本均表现异常.( )",3,{"answer":41,"createTime":5,"id":42,"options":43,"question":48,"source":27,"type":28},[],1081866508,[44,45,46,47],"增加训练数据量","随机翻转训练数据的标签","删除部分训练样本","加密输入数据","在"标签翻转"投毒攻击中,攻击者主要通过什么方式破坏模型?( )",{"answer":50,"createTime":5,"id":51,"options":52,"question":57,"source":27,"type":58},[],1081866509,[53,54,55,56],"墨镜(人脸识别场景)","图像右下角的白色像素点","随机高斯噪声","交通标志上的黄色贴纸","以下哪些是可能的触发器设计?( )",1]