[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fVhs6VbSEJbzcEqUU9x6-0joyPhpd05i6-gle9EldxnE":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":16,"related":17,"source":26,"type":37},[],"2025-04-28 08:42:59",1069534911,[8,9,10,11],"将大语言模型部署在云端服务器上","将大语言模型部署在本地的个人电脑上","积累足够多的个人数据集","重新设计一个全新的语言模型",{"courseId":13,"workId":14,"workName":15},"1000128449","61792908","第十三章单元测试","要发挥私人AI助手的优势,必须满足哪些条件?( )",[18,28,38,47,57,62,65,75],{"answer":19,"createTime":20,"id":21,"options":22,"question":25,"source":26,"type":27},[],"2025-04-28 08:42:58",1069534893,[23,24],"对","错","扩散模型的前向过程是在原始图像经过不断加入噪声生成模糊图像.( )","v2",3,{"answer":29,"createTime":20,"id":30,"options":31,"question":36,"source":26,"type":37},[],1069534894,[32,33,34,35],"将现实世界的实体编码为计算机算法可运算的数据格式","实现文本与图像的匹配","生成高质量的图像","通过训练获得文本编码器和图像编码器的最优参数","CLIP模型的作用和目标有哪些?( )",1,{"answer":39,"createTime":20,"id":40,"options":41,"question":46,"source":26,"type":37},[],1069534899,[42,43,44,45],"它们已经具备了所有专业知识","它们经常会出现答非所问的情况","它们主要是在公开的语料库中训练,缺乏专业知识训练","它们不需要进行微调就能完美适应各种应用场景","关于预训练大语言模型,以下说法正确的是?( )",{"answer":48,"createTime":20,"id":49,"options":50,"question":55,"source":26,"type":56},[],1069534907,[51,52,53,54],"MLLM仅能处理文本信息,无法整合图像等其他模态","MLLM通过整合文本、图像等多种模态信息,可以提供更加全面和丰富的理解和生成能力","MLLM的主要优势在于处理大规模文本数据,与传统文本模型无明显差异","MLLM在处理信息时只关注单一模态,忽视了其他模态的影响","多模态大语言模型(MLLM)具有整合多种模态信息的能力.以下哪项最能准确描述MLLM的优势?( )",0,{"answer":58,"createTime":20,"id":59,"options":60,"question":61,"source":26,"type":27},[],1069534909,[23,24],"大语言模型微调只是在预训练模型基础上进行的简单复制,不涉及迁移学习.( )",{"answer":63,"createTime":5,"id":6,"options":64,"question":16,"source":26,"type":37},[],[8,9,10,11],{"answer":66,"createTime":5,"id":67,"options":68,"question":74,"source":26,"type":37},[],1069534913,[69,70,71,72,73],"提高模型在特定任务上的性能","降低模型的计算复杂度","使模型更好地理解特定领域的知识","增加模型的通用性,以适应更多不同的任务","减少模型的训练时间","微调大语言模型的目的主要包括哪些方面?( )",{"answer":76,"createTime":5,"id":77,"options":78,"question":83,"source":26,"type":56},[],1069534915,[79,80,81,82],"类似人类的耳朵,接收和预处理声学信号","类似人类的大脑,进行理解和推理","类似人类的眼睛,接收和预处理光学信号","类似生成器,输出除文本之外的其他模态","图像和视频编码器在多模态大型语言模型(MLLM)中的作用是什么?( )"]