[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$ffb4EUsBIqpCdqNKuDSdhpMr0wZTTaj-zXMEQTLciAuY":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":16,"related":17,"source":27,"type":28},[],"2025-12-22 08:34:02",278994406,[8,9,10,11],"为所有学习者提供相同难度的标准化内容","只关注语法纠错,不调整对话内容难度","根据学习者水平智能调整对话难度,实现因材施教","仅为高级学习者提供复杂话题讨论",{"courseId":13,"courseImg":14,"courseName":15},"53e1d2ef4961cca8eea3e23969ad2cb9","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F03a579384a6dc297c89809b582fcc767.png","默认课程","AI对话陪练系统在辅助语言学习时,对不同水平学习者采取的策略是?( )",[18,29,38,47,56,65,74,83,86,95],{"answer":19,"createTime":5,"id":20,"options":21,"question":26,"source":27,"type":28},[],278994399,[22,23,24,25],""感知 — 决策 — 行动" 的闭环智能结构",""指令 — 执行" 的单向任务结构",""数据 — 训练 — 预测" 的模型迭代结构",""输入 — 输出" 的简单响应结构","智能体的本质是以下哪种结构?( )","v1",0,{"answer":30,"createTime":5,"id":31,"options":32,"question":37,"source":27,"type":28},[],278994400,[33,34,35,36],"参数量增加到万亿级别","训练成本降低到几千美元","可以并行处理整个句子,且通过自注意力机制直接建立任意两词间的联系","首次实现了中文处理能力","2017年提出的Transformer架构相比RNN的革命性改进是什么?( )",{"answer":39,"createTime":5,"id":40,"options":41,"question":46,"source":27,"type":28},[],278994401,[42,43,44,45],"偏见处理速度快,前见处理速度慢","偏见是恶意的,前见是善意的","偏见基于抽象标签且缺乏修正机制","偏见来自个人经验,前见来自集体文化",""偏见"与"前见"的根本区别在于?( )",{"answer":48,"createTime":5,"id":49,"options":50,"question":55,"source":27,"type":28},[],278994402,[51,52,53,54],"无标签数据","单次预测","静态环境","序列化决策","相较于有监督学习,强化学习的关键区别是什么",{"answer":57,"createTime":5,"id":58,"options":59,"question":64,"source":27,"type":28},[],278994403,[60,61,62,63],"必须先将语音转文字、再将文字转语音才能完成交互","无法继承大语言模型的文本理解与生成能力","能理解说话人语气情绪,并生成带情绪的语音回复","只能处理文本形式的输入数据","与传统语音助手相比,"语音到语音对话模型" 的核心优势是什么?( )",{"answer":66,"createTime":5,"id":67,"options":68,"question":73,"source":27,"type":28},[],278994404,[69,70,71,72],"层次聚类","原型聚类","图聚类","密度聚类","K均值聚类属于哪种聚类方法",{"answer":75,"createTime":5,"id":76,"options":77,"question":82,"source":27,"type":28},[],278994405,[78,79,80,81],"视角鸿沟","尺度鸿沟","语义鸿沟","遮挡鸿沟","计算机中存储的图像是 0-255 数值构成的数字矩阵,与人类看到的富有语义的图像存在差异,这种差异被称为?( )",{"answer":84,"createTime":5,"id":6,"options":85,"question":16,"source":27,"type":28},[],[8,9,10,11],{"answer":87,"createTime":5,"id":88,"options":89,"question":94,"source":27,"type":28},[],278994407,[90,91,92,93],"每个Token的向量是固定不变的,不受上下文影响","向量中的数字是完全随机的,没有实际意义","语义相近的Token,其向量在高维空间中的距离也会比较近","向量维度越高,模型的理解能力就一定越强","在Token化过程中,"你好,世界!"被分解为[你][好][,][世界][!],随后每个Token会转换成高维向量.关于这个向量表示,以下哪个说法最准确?( )",{"answer":96,"createTime":5,"id":97,"options":98,"question":103,"source":27,"type":28},[],278994408,[99,100,101,102],"规模超过千亿的大规模模型参数","仅需短时间训练即可完成模型优化","基于大规模文本数据进行训练","使用超过万亿个词的海量训练数据","大语言模型(如 GPT 系列)能力强大的关键因素不包括以下哪项?( )"]