[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f7HCR72oaWCyFApsiNBHsNaQcDre13tJGLi3IHaHjK48":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":13,"question":17,"related":18,"source":28,"type":47},[],"2025-12-04 18:19:12",1083012017,[8,9,10,11,12],"任务过载(在一个提示中包含过多任务)","引导性偏见(提示的措辞影响模型的公正性)","模糊性(指令不够清晰明确)","隐含假设(默认模型知道某些背景信息)","过度赞美(过多使用表扬性词语)",{"courseId":14,"workId":15,"workName":16},"1000144057","63451123","第八章单元测试","文本中提到了编写提示词时应规避的常见陷阱,其中包括?( )",[19,30,38,48,51],{"answer":20,"createTime":5,"id":21,"options":22,"question":27,"source":28,"type":29},[],1083011907,[23,24,25,26],"从"普通用户"转变为"模型训练师"","从"程序员"转变为"产品经理"","从"对话者"转变为"交互设计师"","从"消费者"转变为"研究员"","文本认为,要进行有效的提示词工程,用户需要完成一次角色转变.这次转变是?( )","v2",0,{"answer":31,"createTime":5,"id":32,"options":33,"question":36,"source":28,"type":37},[],1083011909,[34,35],"对","错","根据文本的"关键洞察",提示词工程的本质在于"创造"能力,一个设计得足够完美的"万能提示词"可以赋予大模型其本身不具备的全新能力.( )",3,{"answer":39,"createTime":5,"id":40,"options":41,"question":46,"source":28,"type":47},[],1083011912,[42,43,44,45],"降低模型对算力的要求","为构建提示词提供了规范、全面的模板","提高了输出结果的稳定性和可控性","使得提示词更具可复用性","使用像 CRISPE、CO-STAR 这样的结构化提示词框架,能带来哪些好处?( )",1,{"answer":49,"createTime":5,"id":6,"options":50,"question":17,"source":28,"type":47},[],[8,9,10,11,12],{"answer":52,"createTime":5,"id":53,"options":54,"question":59,"source":28,"type":29},[],1083012019,[55,56,57,58],"角色扮演 (Role-playing)","提供示例 (Few-shot Learning)","思维链 (Chain of Thought - CoT)","使用分隔符 (Using Delimiters)","根据文本,当处理复杂推理问题时,哪一种提示方法的核心思想是通过引导模型展示其思考步骤来提升表现?( )"]