[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fxtAntnxgcEOYUSrIPxOdS1XpyrsHXytGupJF7NSMMw0":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":13,"question":20,"related":21,"source":32,"type":33},[],"2026-05-23 00:07:33",382646159,[8,9,10,11,12],"遇見","值勤","價值","值得","當、在",{"count":14,"courseId":15,"courseImg":16,"courseName":17,"workId":18,"workName":19},53,"53e1d2ef4961cca8eea3e23969ad2cb9","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F03a579384a6dc297c89809b582fcc767.png","默认课程","work_52079728","第二次作业","適值佗見收.值",[22,34,43,52,61,70,79,88,98,108],{"answer":23,"createTime":24,"id":25,"options":26,"question":31,"source":32,"type":33},[],"2026-04-10 09:36:43",336342297,[27,28,29,30],"词袋模型","规则翻译模型","去噪自编码器","单向语言模型","BART 本质上属于哪一类模型","v1",0,{"answer":35,"createTime":24,"id":36,"options":37,"question":42,"source":32,"type":33},[],336342298,[38,39,40,41],"RoBERTa","T5","GPT","BERT","下列哪一类模型属于 Decoder-only 模型",{"answer":44,"createTime":24,"id":45,"options":46,"question":51,"source":32,"type":33},[],336342299,[47,48,49,50],"SOP 和 NSP","文本生成和机器翻译","MLM 和 NSP","MLM 和 SOP","BERT 的两个核心预训练任务是哪个组合",{"answer":53,"createTime":24,"id":54,"options":55,"question":60,"source":32,"type":33},[],336342300,[56,57,58,59],"对句子做情感分类","预测下一个句子","判断输入中的单词是否被替换过","生成整段文本","在 ELECTRA 中,判别器(Discriminator)的主要任务是什么",{"answer":62,"createTime":24,"id":63,"options":64,"question":69,"source":32,"type":33},[],336342301,[65,66,67,68],"15%","25%","10%","20%","在 BERT 的 Masked Language Model 任务中,输入文本中大约有多少比例的 token 会被 mask",{"answer":71,"createTime":24,"id":72,"options":73,"question":78,"source":32,"type":33},[],336342302,[74,75,76,77],"双向编码、自回归解码","动态掩码、整词掩码","文本填充、文档旋转","词向量因式分解、跨层参数共享","ALBERT 为了减少参数量,主要采用了哪两项技术",{"answer":80,"createTime":24,"id":81,"options":82,"question":87,"source":32,"type":33},[],336342303,[83,84,85,86],"保留 NSP,并减少训练数据","用 LSTM 替代 Transformer","使用动态掩码,并舍弃 NSP","只使用静态掩码,并缩小批次","下列哪一项更符合 RoBERTa 的改进特点",{"answer":89,"createTime":5,"id":90,"options":91,"question":97,"source":32,"type":33},[],382646132,[92,93,94,95,96],"一向","朝向","剛才","以前","接近","向來道邊有賣餅家.向來",{"answer":99,"createTime":5,"id":100,"options":101,"question":107,"source":32,"type":33},[],382646133,[102,103,104,105,106],"將近","并且","大約","尚且","超過","時人以為年且百歲.且",{"answer":109,"createTime":5,"id":110,"options":111,"question":117,"source":32,"type":33},[],382646134,[112,113,114,115,116],"旋即","探索","尋找","古代長度單位,約合一丈","思索","所患尋差.尋"]