[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fcTK4P5Za3N6ciEQhd6CSe6HpM_jUvaxFRYGmRBzffeM":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":16,"related":17,"source":21,"type":22},[],"2026-05-23 11:02:29",383414096,[8,9,10,11],"指令学习","情景学习","思维链","参数微调",{"courseId":13,"courseImg":14,"courseName":15},"53e1d2ef4961cca8eea3e23969ad2cb9","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F03a579384a6dc297c89809b582fcc767.png","默认课程","以下哪项不属于大模型的三个关键能力?( )",[18,23,32,41,50,59,68,77,86,95],{"answer":19,"createTime":5,"id":6,"options":20,"question":16,"source":21,"type":22},[],[8,9,10,11],"v1",0,{"answer":24,"createTime":5,"id":25,"options":26,"question":31,"source":21,"type":22},[],383414097,[27,28,29,30],"通过追问细化需求并明确预期结果","不提供任何背景信息","直接输入初始需求","忽略格式要求","在生成高血压患病因素科普文章时,为获得优质结果的关键步骤是?( )",{"answer":33,"createTime":5,"id":34,"options":35,"question":40,"source":21,"type":22},[],383414098,[36,37,38,39],"刚性物体受力模拟失真","物理因果不可靠","空间与时间细节易混淆","实体计数与持续存在错误","在Sora的失败样例中,当提示"五只灰狼幼崽在碎石路上奔跑追逐"时,出现了"三只幼狼且有一只突然消失后重现"的问题,这属于以下哪类错误?( )",{"answer":42,"createTime":5,"id":43,"options":44,"question":49,"source":21,"type":22},[],383414099,[45,46,47,48],"仅依赖大模型的文本生成能力","直接调用模板引擎生成新格式","通过RAG技术检索相关模板","先解析文档大纲,再调用对应渲染工具","腾讯文档智能助手实现"一键变身"功能(如文档转思维导图)的核心逻辑是?( )",{"answer":51,"createTime":5,"id":52,"options":53,"question":58,"source":21,"type":22},[],383414100,[54,55,56,57],"强化学习","基座训练","有监督微调","预训练","大模型训练流程中,哪个阶段需要使用针对特定任务的标注数据?( )",{"answer":60,"createTime":5,"id":61,"options":62,"question":67,"source":21,"type":22},[],383414101,[63,64,65,66],"利用GAN的生成对抗机制","通过对抗博弈训练生成模型","模拟物理扩散过程的逆过程","直接对噪声进行一步处理生成数据","根据《Sora技术报告》的泛读内容,Diffusion Model通过什么方式实现从噪声到高保真数据的生成?( )",{"answer":69,"createTime":5,"id":70,"options":71,"question":76,"source":21,"type":22},[],383414102,[72,73,74,75],"内存","模型输入输出","回调","数据连接","LangChain中,以下哪个接口用于在链代理调用之间保持状态?( )",{"answer":78,"createTime":5,"id":79,"options":80,"question":85,"source":21,"type":22},[],383414103,[81,82,83,84],"长短期记忆网络(LSTM)","Transformer","卷积神经网络(CNN)","循环神经网络(RNN)","以下哪种架构因并行计算特性对GPU友好,成为大模型基座训练的主流选择?( )",{"answer":87,"createTime":5,"id":88,"options":89,"question":94,"source":21,"type":22},[],383414104,[90,91,92,93],"专注于数据处理速度优化","具备强大共情力","能与用户建立长期连接","拥有独特人设","以下哪个不是角色大模型的核心特点?( )",{"answer":96,"createTime":5,"id":97,"options":98,"question":103,"source":21,"type":22},[],383414105,[99,100,101,102],"Qwen-VL系列","Qwen-VL-Chat","通义万相","Qwen-Omni系列","通义千问多模态模型家族中,以下哪个系列属于全模态模型,能够支持文本、图像、音频、视频高效理解并具备语音生成能力?( )"]