[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fiefw4_mDCj4pDfcRjedieeIyV3f5kaa-4053n8BUG-M":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":19,"related":20,"source":30,"type":31},[],"2025-11-18 14:12:41",236272636,[8,9,10,11],"通过逐像素对比调整图像与文本的像素值以实现对齐","仅增大匹配的图像 - 文本对相似度,不处理不匹配对","增大匹配的图像 - 文本对相似度,同时减小不匹配对的相似度","增大不匹配的图像 - 文本对相似度,同时减小匹配对的相似度",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":18},85,"53e1d2ef4961cca8eea3e23969ad2cb9","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F03a579384a6dc297c89809b582fcc767.png","默认课程","exam_167009830","动手学AI:人工智能通识与实践","CLIP 模型通过 "对比学习" 实现多模态特征对齐时,其核心操作是以下哪一项?( )",[21,32,41,50,59,68,77,86,95,98],{"answer":22,"createTime":5,"id":23,"options":24,"question":29,"source":30,"type":31},[],236272624,[25,26,27,28],"可以训练垂直化的专业领域模型","不需要任何技术门槛就能部署","完全免费使用,没有任何成本","运算速度比云端处理更快","下面哪一项是本地部署AI的优势?( )","v1",0,{"answer":33,"createTime":5,"id":34,"options":35,"question":40,"source":30,"type":31},[],236272625,[36,37,38,39],"ReAct","Plan-and-Solve","预设流程法","静态分步拆解法","以下哪种任务规划方法属于 "动态规划",核心是 "边推理、边行动",能根据环境反馈调整执行路径?( )",{"answer":42,"createTime":5,"id":43,"options":44,"question":49,"source":30,"type":31},[],236272626,[45,46,47,48],"高手经验","随机策略","穷举搜索","数学推导","启发式规则主要来源于",{"answer":51,"createTime":5,"id":52,"options":53,"question":58,"source":30,"type":31},[],236272627,[54,55,56,57],"提高计算速度和存储容量","增加神经网络的层数","完整复制外部世界的所有细节","通过简化表征来辅助信息处理和预测","图灵奖得主杨立昆提出的"内在世界模型"的核心思路是?( )",{"answer":60,"createTime":5,"id":61,"options":62,"question":67,"source":30,"type":31},[],236272628,[63,64,65,66],"9","6","10","8","井字棋博弈树最多层数为",{"answer":69,"createTime":5,"id":70,"options":71,"question":76,"source":30,"type":31},[],236272629,[72,73,74,75],"判别器","印钞机","验钞机","编码器","GAN中"生成器"被比喻成什么",{"answer":78,"createTime":5,"id":79,"options":80,"question":85,"source":30,"type":31},[],236272630,[81,82,83,84],"模型","人","机器","数据","可视分析的目的是将什么放在分析的重要位置中",{"answer":87,"createTime":5,"id":88,"options":89,"question":94,"source":30,"type":31},[],236272633,[90,91,92,93],"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","参数量增加到万亿级别","训练成本降低到几千美元","2017年提出的Transformer架构相比RNN的革命性改进是什么?( )",{"answer":96,"createTime":5,"id":6,"options":97,"question":19,"source":30,"type":31},[],[8,9,10,11],{"answer":99,"createTime":5,"id":100,"options":101,"question":106,"source":30,"type":31},[],236272637,[102,103,104,105],"增加模型复杂度","环境变化时性能稳定","减少数据量","提高训练精度","稳健机器学习的目标是什么"]