第 5/10 份 · 题目 201–250 · 共 452 题 · 点选项后按「确认作答」即刻判对错 · 来源：扫描版 AWS AI 题库（452 题，中英对照）
An education company waftion. The application will give users the ability to enter text or provide a picture of a question. The application will respond with a written answer and an explanation of the written answer. Which model type meets these requirements?
这是一家教育公司的应用程序。该应用能让用户输入文字或上传问题图片。应用会以书面形式给出答案， 并对答案进行解释。的准的业司 哪种型号符合这些要求？
Computer vision model
计算机视觉模型
Large multi-modal language model
大型多模态语言模型
Diffusion model
扩散模型
Text-to-speech model
文本转语音模型
正确答案 B
A large multi-modal language model can natively ingest both text and images as inputs and generate text outputs, making it ideal for a system that accepts typed questions or photos of questions and returns written answers with explanations.
一个大型的多模态语言模型能够原生地接收文本和图像作为输入，并生成文本输出，这使其成为一种理想 的系统，能够接受用户输入的打字问题或问题的照片，并返回带有解释的书面答案。
In which stage of the generative Al model lifecycle are tests performed to examine the model's accuracy?
在生成式人工智能模型的生命周期中，哪个阶段会进行测试以检验模型的准确性？
Deployment
部署
Data selection
数据选择
Fine-tuning
微调
Evaluation
评估
正确答案 D
The evaluation stage is when you run tests and benchmarks, such as accuracy. precision, and other performance metrics, to measure how well the generative Al model performs on hold-out or validation data before moving on to deployment.
评估阶段是在部署之前，您运行测试和基准测试（例如准确性、精确度和其他性能指标），以衡量生 成式人工智能模型在保留数据或验证数据上的表现。
Which statement correctly describes embeddings in generative Al?
以下哪个陈述正确描述了生成式人工智能中的嵌入？
Embeddings represent data as high-dimensional vectors that capture semantic relationships.
嵌入式表示将数据表示为高维向量，从而捕捉语义关系。
Embeddings is a technique that searches data to find the most helpful information to answer natural language questions.
嵌入技术是一种在数据中搜索以找到对回答自然语言问题最有帮助的信息的技术。
Embeddings reduce the hardware requirements of a model by using a less precise data type for the weights and activations.
嵌入技术通过使用精度较低的数据类型来表示权重和激活值，从而降低了模型的硬件需求。
Embeddings provide the ability to store and retrieve data for generative Al applications.
嵌入技术为生成式人工智能应用提供了存储和检索数据的能力。
正确答案 A
Embeddings map inputs like words, sentences, or documents into continuous vector spaces where semantic similarity corresponds to geometric proximity, enabling models to reason about meaning and relationships mathematically.
嵌入将诸如单词、句子或文档之类的输入映射到连续的向量空间，在该空间中，语义相似性对应 于几何上的接近程度，从而使模型能够从数学角度推理意义和关系。
A company wants to add generative Al functionality to its application by integrating a large language model (LLM). The responses from the LLM must be as deterministic and as stable as possible. Which solution meets these requirements?
一家公司希望在其应用程序中添加生成式人工智能功能，方法是集成一个大型语言模型（LLM）。该 语言模型的响应必须尽奇能地确定直稳定。口 哪种解决方案能满足这些要求？
Configure the application to automatically set the temperature parameter to O when submitting the prompt to the LLM.
将应用程序配置为在向LLM提交提示时自动将温度参数设置为0。
Configure the application to automatically add "make your response deterministic" at the end of the prompt before submitting the prompt to the LLM.
将应用程序配置为在向LLM提交提示之前，在提示的末尾自动添加“使您的回复具有确定性”。
Configure the application to automatically add "make your response deterministic" at the beginning of the prompt before submitting the prompt to the LLM.
将应用程序配置为在向LLM提交提示之前，在提示开头自动添加“使您的回复具有确定性”。
Configure the application to automatically set the temperature parameter to 1 when submitting the prompt to the LLM.
将应用程序配置为在向LLM提交提示时自动将温度参数设置为1。
正确答案 A
The temperature hyperparameter directly controls the randomness of the LLM's token sampling: setting it to O forces greedy decoding, yielding the most likely next token every time and thus fully deterministic, stable outputs.
温度超参数直接控制着LLM令牌采样的随机性：将其设置为0会强制采用贪婪解码，每次都选择最 有可能的下一个令牌，从而产生完全确定且稳定的输出。
A company needs to select a generative Al model to build an application. The application must provide responses to users in real time. Which model characteristic should the company consider to meet these requirements?
一家公司需要选择一个生成式人工智能模型来构建一个应用程序。该应用程序必须能够实时为用 户提供响应。 公司应考虑哪种模型特性才能满足这些要求？
Model complexity
模型复杂度
Innovation speed
创新速度
Inference speed
推理速度
Training time
培训时间
正确答案 C
For real-time applications, the model's inference speed - ability to generate responses with low latency - is the critical characteristic to ensure users receive answers promptly.
对于实时应用而言，模型的推理速度一一即生成响应的低延迟能力一一是确保用户能及时获得答案的关键特 性。口
Which term refers to the instructions given to foundation models (FMs) so that the FMs provide a more accurate response to a question?
哪个术语指的是给基础模型（FMs）提供的指令，以便让这些模型对问题给出更准确的回答？
Prompt
提示
Direction
方向
Dialog
对话
Translation
翻译
正确答案 A
A prompt is the input instruction or query you provide to a foundation model to guide its response, ensuring the model understands the task and delivers a more accurate answer.
提示是您提供给基础模型的输入指令或查询，用于引导其响应，确保模型理解任务并给出更准确的答案。
A retail company wants to build an ML model to recommend products to customers. The company wants to build the model based on responsible practices. Which practice should the company apply when collecting data to decrease model bias?
一家零售公司希望构建一个机器学习模型来向客户推荐产品。该公司希望基于负责任的实践来构建该模型。 公司在收集数据以减少模型偏差时应采用哪种做法？
Use data from only customers who match the demographics of the company's overall customer base.
仅使用与公司整体客户群的人口统计特征相匹配的客户的资料。
Collect data from customers who have a past purchase history.
从有过往购买记录的客户那里收集数据。
Ensure that the data is balanced and collected from a diverse group.
确保数据平衡且从多样化的群体中收集。
Ensure that the data is from a publicly available dataset.
确保数据来自公开可用的数据集。
正确答案 C
By gathering training data that reflects the full spectrum of your customer population - across demographics, behaviors, and preferences - you reduce skew toward any one subgroup and help the recommender treat all users equitably.
通过收集能够反映您客户群体全貌的训练数据一一涵盖不同的人口统计特征、行为和偏好一一您可以减 少对任何特定子群体的偏差，帮助推荐系统公平对待所有用户。
A company is developing an ML model to predict customer churn. Which evaluation metric will assess the model's performance on a binary classification task such as predicting churn?
一家公司正在开发一个机器学习模型来预测客户流失。 在诸如预测客户流失这样的二分类任务中，哪种评估指标能够衡量模型的性能？
F1 score
F1分数
Mean squared error (MSE)
均方误差(MSE)
R-squared
决定系数(R²)
Time used to train the model
训练模型所用时间
正确答案 A
For binary classification tasks like churn prediction, the F1 score balances precision and recall into a single metric, making it ideal for evaluating how well the model identifies churners without over- or under-predicting.
对于诸如客户流失预测之类的二分类任务，FI分数将准确率和召回率平衡为一个单一的指标，使其成为评 估模型识别流失客户能力的理想选择，既不会过度预测也不会预测不足。
An Al practitioner is evaluating the performance of an Amazon SageMaker model. The Al practitioner must choose a performance metric. The metric must show the ratio of the number of correctly classified items to the total number of correctly and incorrectly classified items. Which metric meets these requirements?
一位人工智能从业者正在评估亚马逊SageMaker模型的性能。该从业者必须选择一个性能指标。该指标必须 显示正确分类项目数量与正确分类和错误分类项目总数的比率。 哪个指标符合这些要求？
Accuracy
准确性
Precision
精度
F1 score
F1分数
Recall
回忆
正确答案 A
Accuracy is defined as the number of correctly classified items (both true positives and true negatives) divided by the total number of items evaluated (the sum of true positives, true negatives, false positives, and false negatives), matching exactly the ratio described.
准确率被定义为正确分类的项目数量（包括真阳性和真阴性）除以评估的项目总数（真阳性、真阴性、 假阳性和假阴性的总和），与所描述的比例完全一致。
An ecommerce company receives multiple gigabytes of customer data daily. The company uses the data to train an ML model to forecast future product demand. The company needs a solution to perform inferences once each day. Which inference type meets these requirements?
一家电子商务公司每天接收数吉字节的客户数据。该公司利用这些数据训练一个机器学习模型来预测未来 的产品需求。该公司需要一个解决方案，每天进行一次推理。 哪种推理类型符合这些要求？
Batch inference
批量推理
Asynchronous inference
异步推理
Real-time inference
实时推理
Serverless inference
无服务器推理
正确答案 A
Batch inference is designed for high-volume, scheduled predictions: you can point it at the day's gigabytes of data, run a job once daily, and generate all forecasts in one go without needing a persistent endpoint.
批量推理专为高容量、定时预测而设计：您可以将其指向每天的数十亿字节数据，每天运行一次作业，并一 次性生成所有预测结果，而无需持久端点。数缺，
A company has developed a generative Al model for customer segmentation. The model has been deployed in the company's production environment for a long time. The company recently noticed some inconsistency in the model's responses. The company wants to evaluate model bias and drift. WhichAWS service orfeature meets theserequirements?
一家公司开发了一种用于客户细分的生成式人工智能模型。该模型已在公司的生产环境中部署了很长时 间。公司最近注意到该模型的响应存在一些不一致的情况。公司希望评估模型是否存在偏差和漂移。 哪个AWS服务或功能符合这些要求？
Amazon SageMaker Model Monitor
亚马逊SageMaker模型监控器
Amazon SageMaker Clarify
亚马逊 SageMaker Clarify
Amazon SageMaker Model Cards
亚马逊SageMaker模型卡片
Amazon SageMaker Feature Store
亚马逊SageMaker特征存储库
正确答案 A
SageMaker Model Monitor continuously collects inference and ground-truth data to automatically detect data drift, concept drift, and bias shifts in production models. By configuring baseline metrics and thresholds, the company can track inconsistencies in responses over time and receive alerts whenever bias or drift exceed acceptable limits.
SageMaker模型监控器持续收集推理数据和真实数据，以自动检测生产模型中的数据漂移、概念漂移和 偏差变化。通过配置基准指标和阈值，公司可以跟踪响应随时间的变化情况，并在偏差或漂移超出可接 受限度时收到警报。
A company has signed up for Amazon Bedrock access to build applications. The company wants to restrict employeeaccess to specific models available on Amazon Bedrock. Which solution meets these requirements?
一家公司已注册使用亚马逊Bedrock来构建应用程序。该公司希望限制员工对亚马逊Bedrock上特定模型的 访权限。 哪种解决方案能满足这些要求？
Use AWs Identity and Access Management (IAM) policies to restrict model access.
使用AWS身份和访问管理（IAM）策略来限制模型访问权限。
Use AWS Security Token Service (AWS STS) to generate temporary credentials for model use.
使用AWS安全令牌服务（AWSSTS）为模型使用生成临时凭证。
Use AWS Identity and Access Management (IAM) service roles to restrict model subscription.
使用AWS身份和访问管理（IAM）服务角色来限制模型订阅。
Use Amazon Inspector to monitor model access.
使用AmazonInspector监控模型访问情况。
正确答案 A
With IAM policies you can grant or deny Bedrock APl actions scoped to specific model ARNs, ensuring employees can only invoke the exact models you authorize.
借助IAM策略，您可以针对特定模型的ARN授予或拒绝BedrockAPI操作，确保员工只能调用您授权 的确切模型。
Which ML technique uses training data that is labeled with the correct output values?
哪种机器学习技术使用带有正确输出值标签的训练数据？
Supervised learning
监督学习
Unsupervised learning
无监督学习
Reinforcement learning
强化学习
Transfer learning
迁移学习
正确答案 A
Supervised learning trains models on datasets where each example includes both input features and the correct output label, allowing the algorithm to learn the mapping from inputs to known targets.
监督学习是在包含每个示例的输入特征和正确输出标签的数据集上训练模型，从而使算法能够学习从输入 到已知自标的唤射。的证业的效工工
Which large language model (LLM) parameter controls the number of possible next words or tokens considered at each step of the text generation process?
在文本生成过程的每一步中，哪个大型语言模型（LLM）参数控制着所考虑的可能的下一个单词或标 记的数量？
Maximum tokens
最大标记数
Top K
前K项
Temperature
温度
Batch size
批量大小
正确答案 B
The Top K parameter limits token sampling at each generation step to the K most probable next tokens. By choosing how many of the highest-probability candidates to consider, Top K directly controls the breadth of options the model evaluates when generating each token.
TopK参数在每次生成步骤中将标记采样限制为概率最高的K个下一个标记。通过选择考虑多少个概率最高 的候选标记，TopK直接控制了模型在生成每个标记时评估的选项范围。
A company is making a chatbot. The chatbot uses Amazon Lex and Amazon OpenSearch Service. The chatbot uses the company's private data to answer questions. The company needs to convert the data into a vector representation before storing the data in a database. Which type of foundation model (FM) meets these requirements?
一家公司正在开发一个聊天机器人。该聊天机器人使用亚马逊Lex和亚马逊OpenSearch服务。聊天机器人 利用公司的私有数据来回答问题。在将数据存储到数据库之前，公司需要将数据转换为向量表示形式。 哪种类型的基础模型（FM）能满足这些要求？
Text completion model
文本补全模型
Instruction following model
指令遵循模型
Text embeddings model
文本嵌入模型
Image generation model
图像生成模型
正确答案 C
Text embeddings models transform input text into high-dimensional vector representations, making them ideal for storing and retrieving your private data in a vector database for the chatbot.
文本嵌入模型将输入文本转换为高维向量表示，使其成为在向量数据库中存储和检索聊天机器人私有数据 的理想选择。
A company wants to use a large language model (LLM) to generate product descriptions. The company wants to give the model example descriptions that follow a format. Which prompt engineering technique will generate descriptions that match the format?
一家公司希望利用大型语言模型（LLM）生成产品描述。该公司想给模型提供一些遵循特定格式的示例 描述。 哪种提示工程技术能生成符合该格式的描述？?
Zero-shot prompting
零样本提示
Chain-of-thought prompting
思维链提示法
One-shot prompting
一次性提示
Few-shot prompting
少样本提示
正确答案 D
Few-shot prompting provides the model with several example product descriptions that follow your desired format, enabling it to learn the pattern and generate new descriptions in the same style without further training.
少样本提示为模型提供了几个遵循您期望格式的产品描述示例，使模型能够学习这种模式，并在无需进一步 训练的情况下生成相同风格的新描述。
A bank is fine-tuning a large language model (LLM) on Amazon Bedrock to assist customers with questions about their loans. The bank wants to ensure that the model does not reveal any private customer data. Which solution meets these requirements?
一家银行正在亚马逊Bedrock上对一个大型语言模型（LLM）进行微调，以帮助客户解答有关其 贷款的问题。该银行希望确保该模型不会泄露任何客户的私人信息。 哪种解决方案能满足这些要求？
Use Amazon Bedrock Guardrails.
使用亚马逊Bedrock安全防护措施。
Remove personally identifiable information (Pll) from the customer data before fine-tuning the LLM.
在对大型语言模型进行微调之前，从客户数据中移除个人身份信息（PI）。
Increase the Top-K parameter of the LLM.
提高 LLM 的 Top-K参数。
Store customer data in Amazon S3. Encrypt the data before fine-tuning the LLM.
将客户数据存储在亚马逊S3中。在对LLM进行微调之前对数据进行加密。
正确答案 B
Preprocessing your fine-tuning dataset to strip out names, account numbers, and other Pll prevents the model from ever ingesting - or later regurgitating - private customer information. This data sanitization step is the most reliable way to guarantee that sensitive details aren't embedded into the model.
对微调数据集进行预处理，去除姓名、账号及其他个人身份信息，可防止模型获取一一或日后泄露一一客户 的私密信息。这一步数据清理是确保敏感信息不被嵌入模型的最可靠方法。
A grocery store wants to create a chatbot to help customers find products in the store. The chatbot must check the inventory in real time and provide the product location in the store. Which prompt engineering technique should the store use to build the chatbot?
一家杂货店想要创建一个聊天机器人来帮助顾客在店内查找商品。该聊天机器人必须实时检查库存，并 提供商品在店丙的位置。合国合：工协解运位业 该商店应采用哪种提示工程技术来构建聊天机器人？
Zero-shot prompting
零样本提示
Few-shot prompting
少样本提示法
Least-to-most prompting
从最少提示到最多提示
Reasoning and acting (ReAct) prompting
推理与行动（ReAct）提示法
正确答案 D
ReAct prompting interleaves the model's chain-of-thought reasoning with explicit actions" (such as calling your real-time inventory APl), then uses the APl response to inform its final answer. By structuring your prompts to have the LLM think, choose the “Checklnventory" action with the product name, receive the live location data, and then respond to the user, you seamlessly integrate real-time lookups and precise product locations into the chatbot's replies.
ReAct提示法将模型的链式思维推理与明确的“操作”（例如调用您的实时库存API）交替进行，然后使用 API响应来确定其最终答案。通过将提示结构化为让LLM思考、选择带有产品名称的“检查库存”操作、 接收实时位置数据，然后回复用户，您可以将实时查询和精确的产品位置无缝集成到聊天机器人的回复中。
A company uses a third-party model on Amazon Bedrock to analyze confidential documents. The company is concerned about data privacy. Which statement describes how Amazon Bedrock protects data privacy?
一家公司使用亚马逊Bedrock上的第三方模型来分析机密文件。该公司担心数据隐私问题。 以下哪项陈述描述了亚马逊Bedrock如何保护数据隐私？
User inputs and model outputs are anonymized and shared with third-party model providers.
用户输入和模型输出经过名处理后与第三方模型提供商共享。
User inputs and model outputs are not shared with any third-party model providers.
用户输入和模型输出不会与任何第三方模型提供商共享。
User inputs are kept confidential, but model outputs are shared with third-party model providers.
用户输入的内容将予以保密，但模型输出的结果会与第三方模型提供商共享。
User inputs and model outputs are redacted before the inputs and outputs are shared with third-party model providers.
在与第三方模型提供商共享用户输入和模型输出之前，会对这些输入和输出进行脱敏处理。
正确答案 B
Amazon Bedrock processes your data entirely within AWS's secure environment and does not forward your inputs or the generated outputs back to the model vendor, ensuring that your confidential documents remain private.
亚马逊Bedrock完全在AWS的安全环境中处理您的数据，不会将您的输入或生成的输出转发给模型供应商， 从而确保您的机密文件保持私密。
An animation company wants to provide subtitles for its content. Which AWS service meets this requirement?
一家动画公司想要为其内容提供字幕。 哪个AWS服务能满足这一要求？
Amazon Comprehend
亚马逊comprehend服务
Amazon Polly
亚马逊波利（Amazon Polly）
Amazon Transcribe
亚马逊转录服务
Amazon Translate
亚马逊翻译
正确答案 C
Amazon Transcribe converts spoken audio into accurate, time-stamped text transcripts, making it the ideal service for generating subtitles from your animation audio tracks.
亚马逊Transcribe能将语音音频转换为准确的时间戳文本记录，是为您的动画音频轨道生成字幕的理想 服务。
An ecommerce company wants to group customers based on their purchase history and preferences to personalize the user experience of the company's application. Which ML technique should the company use?
一家电子商务公司希望根据客户的购买历史和偏好对客户进行分组，以个性化其应用程序的用户体验。 公司应该使用哪种机器学习技术？
Classification
分类
Clustering
聚类
Regression
回归分析
Content generation
内容生成
正确答案 B
Clustering is an unsupervised learning technique that automatically groups data points - in this case, customers with similar purchase histories and preferences - into segments without needing predefined labels, enabling personalized experiences.
聚类是一种无监督学习技术，它能自动将数据点（在此例中为具有相似购买历史和偏好的客户）分 组到不同的细分市场，无需预先定义标签，从而实现个性化体验。
A company wants to control employee access to publicly available foundation models (FMs). Which solution meets these requirements?
一家公司希望控制员工对公开可用的基础模型（FMs）的访问权限。 哪种解决方案能满足这些要求？
Analyze cost and usage reports in AWS Cost Explorer.
在AWS成本管理器中分析成本和使用情况报告。
Download AWS security and compliance documents from AWS Artifact.
从AWSArtifact下载AWS安全和合规性文档。
Configure Amazon SageMaker JumpStart to restrict discoverable FMs.
配置Amazon SageMakerJumpStart以限制可发现的预训练模型。
Build a hybrid search solution by using Amazon OpenSearch Service.
通过使用亚马逊OpenSearch服务构建混合搜索解决方案。
正确答案 C
SageMaker JumpStart lets you curate which publicly available foundation models appear in your account's model catalog. By using JumpStart's model catalog settings (in combination with IAM), you can prevent employees from discovering or deploying any FMs you haven't explicitly allowed.
SageMakerJumpStart让您能够筛选出哪些公开可用的基础模型会出现在您账户的模型目录中。通过使用 JumpStart的模型目录设置（结合IAM），您可以防止员工发现或部署您未明确允许的任何基础模型。
A company has set up a translation tool to help its customer service team handle issues from customers around the world. The company wants to evaluate the performance of the translation tool. The company sets up a parallel data process that compares the responses from the tool to responses from actual humans. Both sets of responses are generated on the same set of documents. Which strategy should the company use to evaluate the translation tool?
一家公司已开发出一款翻译工具，旨在协助其客服团队处理来自世界各地客户的各类问题。该公司希 望评估这款翻译工具的性能。为此，公司设立了平行数据处理流程，将翻译工具生成的回复与人工回 复进行对比。两组回复均基于同一套文档生成。 公司应该采用哪种策略来评估翻译工具？
Use the Bilingual Evaluation Understudy (BLEU) score to estimate the absolute translation quality of the two methods.
使用双语评估替补（BLEU）分数来估计这两种方法的绝对翻译质量。
Use the Bilingual Evaluation Understudy (BLEU) score to estimate the relative translation quality of the two methods.
使用双语评估替补（BLEU）分数来估计这两种方法的相对翻译质量。
Use the BERTScore to estimate the absolute translation quality of the two methods.
使用BERTScore来评估这两种方法的绝对翻译质量。
Use the BERTScore to estimate the relative translation quality of the two methods.
使用BERTScore来评估这两种方法的相对翻译质量。
正确答案 B
BLEU measures n-gram overlap between a candidate translation and one or more reference translations. When you apply it to both the tool's outputs and the human outputs on the same source documents, you get comparable scores that let you directly assess which approach yields better translations.
BLEU通过计算候选译文与一个或多个参考译文之间的n元语法重叠来衡量二者的一致性。当您将其应用 于工具输出和人工翻译的相同源文档时，您会得到可比的分数，从而能够直接评估哪种方法能产生更好的 译文。
An Al practitioner wants to generate more diverse and more creative outputs from a large language model (LLM). How should the Al practitioner adjust the inference parameter?
一位人工智能从业者希望从大型语言模型（LLM）中生成更多样化、更具创意的输出结果。 人工智能从业者应如何调整推理参数？
Increase the temperature value.
提高温度值。
Decrease the Top K value.
减小TopK值。
Increase the response length.
增加回复长度。
Decrease the prompt length.
减少提示长度。
正确答案 A
Raising the temperature softens the model's probability distribution, increasing the chance of sampling less-likely tokens and thus producing more varied and creative responses.
提高温度会使模型的概率分布变软，增加采样不太可能的标记的机会，从而生成更多样化和更具创意的 响应。
A company has developed custom computer vision models. The company needs a user-friendly interface for data labeling to minimize model mistakes on new real-world data. Which AWS service, feature, or tool meets these requirements?
一家公司开发了定制的计算机视觉模型。该公司需要一个用户友好的数据标注界面，以尽量减少模型在新 的实际数据上的错误。 哪种AWS服务、功能或工具能满足这些要求？
Amazon SageMaker Ground Truth
亚马逊 SageMaker Ground Truth
Amazon SageMaker Canvas
亚马逊 SageMaker Canvas
Amazon Bedrock playground
亚马逊Bedrock沙盒
Amazon Bedrock Agents
亚马逊Bedrock代理
正确答案 A
SageMaker Ground Truth provides a built-in, easy-to-use labeling console with workflows and quality controls specifically for computer vision tasks, enabling you to efficiently generate high-quality labeled data and reduce model errors on real-world images.
SageMakerGroundTruth提供了一个内置且易于使用的标注控制台，其中包含专门针对计算机视觉任务的工 作流程和质量控制功能，使您能够高效生成高质量的标注数据，并减少模型在真实世界图像上的错误。
A company is integrating Al into its employee recruitment and hiring solution. The company wants to mitigate bias risks and ensure responsible Al practices while prioritizing equitable hiring decisions. Which core dimensions of responsible Al should the company consider? (Choose two.)
一家公司正在将其员工招聘和录用解决方案与人工智能相结合。该公司希望降低偏见风险，确保负责 任的人工智能实践，同时优先考虑公平的招聘决策。 公司应考虑负责任的人工智能的哪些核心维度？（选择两个。）
Fairness
公平性
Tolerance
容忍度
Flexibility
灵活性
Open source
开源
Transparency
透明度
正确答案 A,E
Fairness - Ensuring equitable outcomes across all demographic groups helps mitigate bias risks in hiring decisions. Transparency - Providing clear explanations of how the Al makes decisions builds trust and allows auditing for bias.
公平性一一确保所有人口群体都能获得公平的结果有助于降低招聘决策中的偏见风险。 透明度一一清晰解释人工智能如何做出决策有助于建立信任，并允许对偏见进行审计。
A financial company has deployed an ML model to predict customer churn. The model has been running in production for 1 week. The company wants to evaluate how accurately the model predicts churn compared to actual customer behavior. Which metric meets these requirements?
一家金融公司部署了一个机器学习模型来预测客户流失情况。该模型已在生产环境中运行了一周。该公司希 望评估该模型预测客户流失的准确性与实际客户行为之间的差异。 哪个指标符合这些要求？
Root mean squared error (RMSE)
均方根误差(RMSE)
Return on investment (ROI)
投资回报率(ROI)
F1 score
F1 分数
Bilingual Evaluation Understudy (BLEU) score
双语评估参考（BLEU）分数
正确答案 C
The F1 score combines precision (how many predicted churners actually churned) and recall (how many actual churners were correctly identified) into a single metric, making it ideal for measuring your model's accuracy on the binary churn - prediction task.
F1分数将准确率（预测流失的客户中实际流失的比例）和召回率（实际流失的客户中被正确识别的比例) 结合为一个指标，使其成为衡量二元流失预测任务中模型准确性的理想选择。
A company has a generative Al application that uses a pre-trained foundation model (FM) on Amazon Bedrock. The company wants the FM to include more context by using company information. Which solution meets these requirements MOsT cost-effectively?
一家公司拥有一款基于亚马逊Bedrock上的预训练基础模型（FM）的生成式人工智能应用程序。该公 司希望该基础模型能纳入更多公司信息，从而提供更丰富的上下文。 哪种解决方案能以最经济高效的方式满足这些要求？
Use Amazon Bedrock Knowledge Bases.
使用亚马逊Bedrock知识库。
Choose a different FM on Amazon Bedrock.
在亚马逊Bedrock上选择一个不同的FM。
Use Amazon Bedrock Agents.
使用亚马逊Bedrock代理。
Deploy a custom model on Amazon Bedrock.
在亚马逊Bedrock上部署自定义模型。
正确答案 A
Amazon Bedrock Knowledge Bases let you index and retrieve your company's documents at inference time, injecting relevant context into the pre-trained FM's prompts without the cost or complexity of re-training or fine-tuning a custom model. This retrieval-augmented approach delivers up-to-date, domain-specific information in responses while keeping costs low.
亚马逊Bedrock知识库让您能够在推理时索引和检索公司的文档，在预训练的FM提示中注入相关上下 文，而无需承担重新训练或微调自定义模型的成本和复杂性。这种检索增强的方法在响应中提供最新的 特定领域信息，同时保持成本低廉。
HOTSPOT A company is using Amazon SageMaker to develop Al models. Select the correct SageMaker feature or resource from the following list for each step in the Al model lifecycle workflow. Each SageMaker feature or resource should be selected one time or not at all. Hot Area: Answer Area Managingdifferentversions ofthemodel ggacsosoo SageMakerClarify SageMakerModel Beaistry SageMakerModelRegistry SageMakerServerlessInference agemaker Serveness erence Using the currentmodeIto makepredictions Scaccoapccli SageMakerClarify SageMakerModelBegistry SageMakerModelRegistry SageMakerServerlessInference agemakerServeness merece
热点 一家公司正在使用亚马逊的SageMaker来开发人工智能模型。 从以下列表中为AI模型生命周期工作流中的每个步骤选择正确的SageMaker功能或资源。每个 SageMaker功能或资源只能选择一次或根本不选。 热点区域： Answer Area Managingdifferentversionsof themodel SageMaker Clarify SageMakerModel Registry SageMakerServerless Inference Usingthecurrentmodeltomakepredictions SageMaker Clarify SageMakerModel Registry SageMaker Serverless Inference
拖拽匹配题（HOTSPOT）：考试中需从下拉列表为每一项匹配答案，此处仅作题目与答案解析的展示复习。
此题答案见下方解析
A food service company wants to collect a dataset to predict customer food preferences. The company wants to ensure that the food preferences of all demographics are included in the data. Which dataset characteristic does this scenario present?
一家餐饮服务公司想要收集一个数据集来预测顾客的食品偏好。该公司希望确保数据涵盖所有人群的食 品偏好。 这种情况呈现出哪种数据集特征？
Accuracy
准确性
Diversity
多样性
Recency bias
近因效应偏差
Reliability
可靠性
正确答案 B
Ensuring representation of all demographic groups addresses the dataset's diversity, so the model learns from a wide range of customer preferences.
确保所有人口群体都有代表性，这解决了数据集的多样性问题，从而使模型能够从广泛的客户偏好中学习。
A company wants to create a chatbot that answers questions about human resources policies. The company is using a large language model (LLM) and has a large digital documentation base. Which technique should the company use to optimize the generated responses?
一家公司想要开发一个能回答有关人力资源政策问题的聊天机器人。该公司正在使用一个大型语言 模型（LLM），并且拥有一个庞大的数字化文档库。 公司应该采用哪种技术来优化生成的回复？
Use Retrieval Augmented Generation (RAG).
使用检索增强生成（RAG）。
Use few-shot prompting.
使用少样本提示。
Set the temperature to 1.
将温度设定为 1,
Decrease the token size.
减小标记大小。
正确答案 A
RAG lets the chatbot pull in precise, up-to-date passages from your HR documentation at inference time, grounding its answers in the actual policy text and ensuring accuracy without overloading the LLM's context window.
检索增强生成（RAG）能让聊天机器人在推理时从您的人力资源文档中提取精确且最新的段落，其回答基 于实际的政策文本，从而确保准确性，同时不会使大型语言模型（LLM）的上下文窗口过载。
An education company is building a chatbot whose target audience is teenagers. The company is training a custom large language model (LLM). The company wants the chatbot to speak in the target audience's language style by using creative spelling and shortened words. Which metric will assess the LLM's performance?
一家教育公司正在开发一款面向青少年的聊天机器人。该公司正在训练一个定制的大语言模型（LLM）。 公司希望聊天机器人能以青少年的语言风格进行交流，使用创意拼写和缩写词。 将用哪个指标来评估大语言模型的性能？
F1 score
F1分数
BERTScore
BERTScore
Recall-Oriented Understudy for Gisting Evaluation (ROUGE)
基于召回率的摘要评估工具（ROUGE)
Bilingual Evaluation Understudy (BLEU) score
双语评估参考（BLEU）分数
正确答案 D
BLEU evaluates surface-level n-gram overlap between the LLM's outputs and reference examples--in this case, targets written in creative spelling and shorthand, making it well suited to measure how closely the model's style matches the teenager-oriented language.
BLEU评估的是语言模型输出内容与参考示例（在这种情况下是采用创意拼写和缩写的文本）在表面层面的 n元语法重叠情况，这使其非常适合衡量模型的风格与面向肯少年的语言的接近程度。
A customer service team is developing an application to analyze customer feedback and automatically classify the feedback into different categories. The categories include product quality, customer service, and delivery experience. Which Al concept does this scenario present?
一个客户服务团队正在开发一款应用程序，用于分析客户反馈，并自动将反馈归入不同的类别。这些类 别包括产品质量、客户服务和配送体验。 这个场景呈现了哪种人工智能概念？
Computer vision
计算机视觉
Natural language processing (NLP)
自然语言处理（NLP)
Recommendation systems
推荐系统
Fraud detection
诈骗检测
正确答案 B
Automatically analyzing and classifying free-text feedback into thematic categories is a core NLP task (text classification).
自动将自由文本反馈分析并分类到主题类别中是一项核心的自然语言处理任务（文本分类）。
A financial services company must ensure that its generative Al-powered chatbot provides factual responses for regulatory compliance. Which solution prevents the underlying foundation model (FM) from hallucinating?
一家金融服务公司必须确保其由生成式人工智能驱动的聊天机器人所提供的回答准确无误，以符合 监管要求。工品司) 哪种解决方案能防止基础模型（FM）产生幻觉？
Use AWS Config to query compliance metadata by using natural language.
使用AWSConfig通过自然语言查询合规性元数据。
Configure Amazon Bedrock Guardrails to evaluate user inputs and model responses.
配置亚马逊Bedrock安全防护措施，以评估用户输入和模型响应。
Use Amazon Fraud Detector to detect potentially fraudulent online activities.
使用亚马逊欺诈检测器来检测潜在的欺诈性在线活动。
Use AWS Audit Manager to prepare IT audit and compliance reports.
使用AWS审计管理器准备IT审计和合规报告。
正确答案 B
Amazon Bedrock Guardrails allows you to define rules and policies that evaluate and constrain both user inputs and model responses. By configuring guardrails, you can prevent the generative Al model from producing hallucinated or non-factual outputs, ensuring the chatbot adheres to compliance requirements and provides factual responses. This directly addresses the need to control hallucinations in generative Al systems for regulatory compliance.
亚马逊Bedrock安全护栏功能允许您定义规则和策略，以评估和限制用户输入以及模型响应。通过配置安 全护栏，您可以防止生成式人工智能模型生成虚构或不实的输出，确保聊天机器人符合合规要求并提供真 实准确的响应。这直接解决了在生成式人工智能系统中控制虚构内容以满足监管合规性的需求。
HOTSPOT A company wants to develop a solution that uses generative Al to create content for product advertisements, including sample images and slogans. Select the correct model type from the following list for each action. Each model type should be selected one time. Hot Area: Create high-quality images that are influenced by the generated slogans and product Select...-- aogqaanyagooaramaonooapyagocraoaooganoanapioaaorBeic... Diffusion model _ ! Object detection model ojeeforeechorodmedel Transformer-based model Create contextually relevant slogans based on theadvertisementproductSelect... oogoaaoopoec Diffusion model Object detectionmodel Tropeformechosedmedel Transformer-based model Ensurethatcompanybrandelements areproperlyplaced in theimagesSelect... Ehsrenatcompanybranacicesarepropenypiaccaemageserectda Diffusion model Object detection model Transformerbasedmodel Transtormer-based model
热点 一家公司希望开发一种解决方案，利用生成式人工智能为产品广告创作内容，包括示例图片 和产告语。 从以下列表中为每个动作选择正确的模型类型。每个模型类型应选择一次。 热点区域： Create high-quality images that are influenced by the generated slogans and productSelect... Diffusion model . ! Object detection model Transformer-based model Create contextuallyrelevant slogans based on the advertsement productSelect.. Diffusion model object detection model Transformer-based model Ensure that company brand elements are properlyplaced in the imagesSelect.. Diffusion model Object detection model Transformer-based model
拖拽匹配题（HOTSPOT）：考试中需从下拉列表为每一项匹配答案，此处仅作题目与答案解析的展示复习。
此题答案见下方解析
Create high-quality images that are influenced by the generated slogans and product -> Diffusion model Create contextually relevant slogans based on the advertisement product -> Transformer-based model Ensure that company brand elements are properly placed in the images -> Object detection model Diffusion models are state-of-the-art for high-quality image generation. Transformer-based models excel at natural language processing, making them suitable for generating contextually relevant slogans. Object detection models are designed to identify and ensure correct placement of specific objects (e.g., brand elements) within images.
生成受标语和产品启发的高质量图像->扩散模型基于广告产品生成相关标语->基于Transformer的模 型确保公司品牌元素在图像中正确放置->目标检测模型 扩散模型是高质量图像生成的前沿技术。 基于Transformer的模型在自然语言处理方面表现出色，因此非常适合生成与上下文相关的标语。 对象检测模型旨在识别图像中特定对象（例如品牌元素）并确保其正确放置。
A company has created multiple ML models. The company needs a solution for storing, managing, and versioning the models. Which AWS service or feature meets these requirements?
一家公司已创建了多个机器学习模型。该公司需要一个解决方案来存储、管理和对这些模型进行版本 控制。 哪个AWS服务或功能符合这些要求？
AWS Audit Manager
AWS审计管理器
Amazon SageMaker Model Monitor
亚马逊SageMaker模型监控器
Amazon SageMaker Model Registry
亚马逊SageMaker模型注册表
Amazon SageMaker Canvas
亚马逊 SageMaker Canvas
正确答案 C
Amazon SageMaker Model Registry is the AWS service designed to store, manage, and version machine learning models. It provides a central repository for tracking model versions and managing their deployment status throughout theML lifecycle.
亚马逊SageMaker模型注册中心是AWS提供的一项服务，旨在存储、管理和版本控制机器学习模型。 它提供了一个中央存储库，用于跟踪模型版本，并在整个机器学习生命周期中管理其部署状态。
An Al practitioner is building an ML model. The Al practitioner wants to provide model transparency and explainability to stakeholders. Which solution will meet these requirements?
一位人工智能从业者正在构建一个机器学习模型。该从业者希望向利益相关者提供模型的透明度和可 解释性。 哪种解决方案能满足这些要求？
Present the model Shapley values.
展示模型的夏普利值。
Provide the model accuracy measure.
提供模型的准确度衡量指标。
Provide the model confusion matrix.
提供模型混淆矩阵。
Provide a secure model inference endpoint.
提供一个安全的模型推理端点。
正确答案 A
Presenting the model Shapley values provides transparency and explainability by showing how each feature contributes to individual predictions, helping stakeholders understand the reasoning behind the model's outputs.
展示模型的夏普利值能够提供透明度和可解释性，通过展示每个特征对单个预测的贡献，帮助利益相 关者理解模型输出背后的逻辑。
A company is developing an ML application. The application must automatically group similar customers and products based on their characteristics. Which ML strategy should the company use to meet these requirements?
一家公司正在开发一个机器学习应用程序。该应用程序必须根据客户和产品的特征自动将相似的客户 和产品进行分组。为了满足这些要求，该公司应采用哪种机器学习策略？
Unsupervised learning
无监督学习
Supervised learning
监督学习
Reinforcement learning
强化学习
Semi-supervised learning
半监督学习
正确答案 A
Unsupervised learning is used to automatically group or cluster similar customers and products based on their characteristics without the need for labeled data. This strategy fits scenarios where the goal is to discover patterns or groupings in the data.
无监督学习用于根据客户和产品的特征自动对其进行分组或聚类，无需使用有标签的数据。这种策略适 用于旨在发现数据中的模式或分组的场景。
A news agency publishes articles in English. The agency wants to make articles available in other languages. Which solution meets these requirements?
一家新闻机构用英语发布文章。该机构希望将文章提供给其他语言的读者。 哪种解决方案能满足这些要求？
Add Amazon Transcribe to the company's website.
在公司的网站上添加亚马逊转录服务。
Use the Amazon Translate real-time translation feature.
使用亚马逊翻译的实时翻译功能。
Add Amazon Personalize to the company's website.
将亚马逊个性化服务添加到公司的网站上。
Use the Amazon Textract real-time document processing feature.
使用AmazonTextract的实时文档处理功能。
正确答案 B
Amazon Translate provides real-time translation of text, making it suitable for automatically translating articles from English into other languages for publication.
亚马逊翻译服务能够实时翻译文本，因此非常适合将英文文章自动翻译成其他语言进行发布。
A bank is building a chatbot to answer customer questions about opening a bank account. The chatbot will use public bank documents to generate responses. The company will use Amazon Bedrock and prompt engineering to improve the chatbot's responses. Which prompt engineering technique meets these requirements?
一家银行正在开发一个聊天机器人来回答客户关于开设银行账户的问题。该聊天机器人将利用公开的银 行文件生成回复。该公司将使用亚马逊Bedrock和提示工程来改进聊天机器人的回复。 哪种提示工程技术能满足这些要求？
Complexity-based prompting
基于复杂性的提示
Zero-shot prompting
零样本提示法
Few-shot prompting
少样本提示
Directional stimulus prompting
方向性刺激提示
正确答案 C
Few-shot prompting involves providing the generative Al model with several examples (drawn from public bank documents) to guide it in generating more accurate and relevant responses. This technique helps the chatbot better align its answers with the desired content and format.
少样本提示是指向生成式人工智能模型提供几个示例（这些示例来自公开的银行文件），以引导其生成 更准确和相关的回复。这种技术有助于聊天机器人更好地使其答案与期望的内容和格式保持一致。
A company wants to fine-tune an ML model that is hosted on Amazon Bedrock. The company wants to use its own sensitive data that is stored in private databases in a VPC. The data needs to stay within the company's private network. Which solution will meet these requirements?
一家公司希望对托管在亚马逊Bedrock上的机器学习模型进行微调。该公司希望使用存储在VPC中私 有数据库里的自身敏感数据。这些数据必须留在公司的私有网络内。 哪种解决方案能满足这些要求？
Restrict access to Amazon Bedrock by using an AWS Identity and Access Management (IAM) service role.
使用AWS身份和访问管理（IAM）服务角色来限制对AmazonBedrock的访问。
Restrict access to Amazon Bedrock by using an AWS Identity and Access Management (IAM) resource policy.
使用AWS身份和访问管理（IAM）资源策略限制对AmazonBedrock的访问。
Use AWS PrivateLink to connect the VPC and Amazon Bedrock.
使用AWSPrivateLink将VPC与AmazonBedrock连接起来。
Use AWS Key Management Service (AWS KMS) keys to encrypt the data.
使用AWS密钥管理服务（AWSKMS）密钥对数据进行加密。
正确答案 C
AWS PrivateLink enables secure, private connectivity between your VPC and Amazon Bedrock without exposing data to the public internet, ensuring that sensitive data stays within your private network during fine-tuning.
AWSPrivateLink可实现您的虚拟私有云（VPC）与AmazonBedrock之间的安全私有连接，无需将数 据暴露于公共互联网，从而确保在微调过程中敏感数据始终处于您的私有网络内。
A documentary filmmaker wants to reach more viewers. The filmmaker wants to automatically add subtitles and voice-overs in multiple languages to their films. Which combination of steps will meet these requirements? (Choose two.)
一位纪录片导演希望吸引更多观众。该导演希望为其影片自动添加多种语言的字幕和旁白。 哪两种步骤组合能满足这些要求？（选择两项。）
Use Amazon Transcribe and Amazon Translate to generate subtitles in other languages.
使用亚马逊转录服务（AmazonTranscribe）和亚马逊翻译服务（AmazonTranslate）生成其他语言的字幕。
Use Amazon Textract and Amazon Translate to generate subtitles in other languages.
使用亚马逊Textract和亚马逊Translate生成其他语言的字幕。
Use Amazon Polly to generate voice-overs in other languages.
使用亚马逊Polly生成其他语言的旁白。
Use Amazon Translate to generate voice-overs in other languages.
使用亚马逊翻译生成其他语言的旁白。
Use Amazon Textract to generate voice-overs in other languages.
使用亚马逊Textract生成其他语言的旁白。
正确答案 A,C
Use Amazon Transcribe and Amazon Translate to generate subtitles in other languages: Amazon Transcribe converts spoken dialogue to text (subtitles), and Amazon Translate can then translate these subtitles into multiple languages. Use Amazon Polly to generate voice-overs in other languages: Amazon Polly converts translated text into lifelike speech, enabling the creation of multilingual voice-overs.
使用亚马逊转录服务（AmazonTranscribe）和亚马逊翻译服务（AmazonTranslate）生成其他语言的 字幕：亚马逊转录服务将对话内容转换为文本（字幕），然后亚马逊翻译服务可以将这些字幕翻译成 多种语言。 使用亚马逊Polly生成其他语言的旁白：亚马逊Polly可将翻译后的文本转换为逼真的语音，从而能够 制作多语言旁百。主
A company wants to create a chatbot to answer employee questions about company policies. Company policies are updated frequently. The chatbot must reflect the changes in near real time. The company wants to choose a large language model (LLM). Which solution meets these requirements?
一家公司希望创建一个聊天机器人来回答员工有关公司政策的问题。公司政策经常更新。该聊天机器人 必须近乎实时地反映这些变化。该公司希望选择一个大型语言模型（LLM）。 哪种解决方案能满足这些要求？
Fine-tune an LLM on the company policy text by using Amazon SageMaker.
使用亚马逊SageMaker在公司政策文本上对大型语言模型进行微调。
Select a foundation model (FM) from Amazon Bedrock to build an application.
从亚马逊Bedrock中选择一个基础模型（FM）来构建应用程序。
Create a Retrieval Augmented Generation (RAG) workflow by using Amazon Bedrock Knowledge Bases.
利用业马逊Bedrock知识库创建检索增强生成（RAG）工作流。
Use Amazon Q Business to build a custom Q App.
使用亚马逊Q商业版构建自定义Q应用程序。
正确答案 C
A RAG workflow with Amazon Bedrock Knowledge Bases allows the chatbot to access the most recent company policy documents dynamically. ensuring responses reflect policy updates in near real time without the need to retrain or fine-tune the LLM each time content changes.
使用亚马逊Bedrock知识库的RAG工作流使聊天机器人能够动态访问最新的公司政策文件，确保响应能 够近乎实时地反映政策更新，而无需每次内容变更时都重新训练或微调大语言模型。
A company is using supervised learning to train an Al model on a small labeled dataset that is specific to a target task. Which step of the foundation model (FM) lifecycle does this describe?
一家公司正在使用监督学习，在一个针对特定任务的小型标注数据集上训练一个人工智能模型。 这描述的是基础模型（FM）生命周期中的哪一步？
Fine-tuning
微调
Data selection
数据选择
Pre-training
预训练
Evaluation
评估
正确答案 A
Fine-tuning is the process of training a pre-existing foundation model on a smaller, labeled dataset that is specific to a target task, allowing the model to adapt to the company's requirements.
微调是指在较小的、针对特定任务的有标签数据集上训练一个已有的基础模型，从而使该模型能够适应公 司的需求。
HOTSPOT A company is developing an Al application to help the company approve or deny personal loans. The application must follow the principles of responsible Al. Select the correct responsible Al principle from the following list for each action. Select each responsible Al principle one time or not at all. Hot Area: Encrypt the application data,and isolate the application on a private network.Selec.. Select. Explainability Pavaes and security Robustness secuty Safety Evaluate how dfferent populationgroups will be impacted.Select.. Select. Explainability Paivaes and security Robustness secunty Safety Test the application with unexpected data to ensure the application will work in unique situations.Select... Select. Explainability Paivaes and security Robustness secuty Safety
热点 一家公司正在开发一款人工智能应用程序，以帮助公司批准或拒绝个人贷款。该应用程序必须遵循负 责任的人工智能原则。 从以下列表中为每个行动选择正确的负责任的人工智能原则。每个负责任的人工智能原则只能选择一次或 根本不选。 热点区域： Encrypt the application data and isolate the application on a private network.Select.. pSelect. Explainability Fairness Privacy and security Robustness Safety Evaluate how different population groups willbe impacted.Select.. Select... Explainability Fairness Privacy and security Robustness Safety Test the application with unexpected data to ensure the application will work in unique situations.Select.. Select.. Explainability Fairness Privacy and security Robustness Safety
拖拽匹配题（HOTSPOT）：考试中需从下拉列表为每一项匹配答案，此处仅作题目与答案解析的展示复习。
此题答案见下方解析
Encrypt the application data, and isolate the application on a private network -> Privacy and security Evaluate how different population groups will be impacted -> Fairness Test the application with unexpected data to ensure the application will work in unique situations -> Robustness Protecting data and networks is part of privacy and security. Assessing impacts on different groups ensures fairness. Testing with unexpected data checks the application's robustness.
对应用程序数据进行加密，并将应用程序隔离在专用网络中>隐私与安全评估不同人群将受到的 影响>公平性“ 使用意外数据测试应用程序，以确保其在特殊情况下仍能正常运行>稳健性 保护数据和网络是隐私和安全的一部分。 评估对不同群体的影响有助于确保公平。 使用意外数据进行测试可以检验应用程序的健壮性。
A company is introducing a new feature for its application. The feature will refine the style of output messages. The company will fine-tune a large language model (LLM) on Amazon Bedrock to implement the feature. Which type of data does the company need to meet these requirements?
一家公司正在为其应用程序引入一项新功能。该功能将优化输出消息的风格。该公司将在亚马逊 Bedrock上微调一个大型语言模型（LLM）来实现这一功能。 公司需要哪种类型的数据才能满足这些要求？
Samples of only input messages
仅输入消息的样本
Samples of only output messages
仅输出消息的示例
Samples of pairs of input and output messages
输入消息与输出消息成对的示例
Separate samples of input and output messages
分别提供输入和输出消息的样本
正确答案 C
Fine-tuning a large language model to refine the style of output messages requires training data consisting of paired input and output messages. These pairs allow the model to learn the relationship between the input provided and the desired styled output, ensuring the fine-tuned model produces responses with the intended style.
微调大型语言模型以优化输出消息的风格，需要包含成对的输入和输出消息的训练数据。这些成对的数据能 让模型学习输入内容与期望风格的输出之间的关系，从而确保微调后的模型能够生成具有预期风格的回复。
A healthcare company is building an Al solution to predict patient readmission within 30 days of patient discharge. The company has trained a model on historical patient data including medical history, demographics, and treatment specifications, to provide readmission predictions in real time. Which task describes Al model inference in this scenario?
一家医疗保健公司正在开发一种人工智能解决方案，用于预测患者出院后30天内的再入院情况。该公 司利用包括病史、人口统计信息和治疗详情在内的历史患者数据训练子一个模型，以实时提供再入院预 测。 在这个场景中，哪项任务描述了人工智能模型的推理过程？
Gather historical patient readmission data.
收集患者历史再入院数据。
Use appropriate metrics and assess model performance.
使用适当的指标并评估模型性能。
Use data to identify patient patterns and correlations.
利用数据识别患者模式和关联。
Use a trained model to predict patient readmission.
使用训练好的模型来预测患者的再次入院情况。
正确答案 D
Model inference refers to the process of applying a trained model to new data--in this case, using the trained model to make real-time predictions about whether a patient will be readmitted within 30 days of discharge.
模型推理指的是将训练好的模型应用于新数据的过程一一在此情况下，即利用训练好的模型对患者出院 后30天内是否会被再次收治入院进行实时预测。
A financial company wants to build workflows for human review of ML predictions. The company wants to define confidence thresholds for its use case and adjust the thresholds over time. Which AWS service meets these requirements?
一家金融公司希望构建用于人工审核机器学习预测的工作流程。该公司希望为其应用场景定义置信度阈值， 并随著时简推移对这些阈值进行调整。量国 哪种AWS服务能满足这些要求？
AmazonPersonalize
亚马逊个性化服务
Amazon Augmented Al (Amazon A21)
亚马逊增强型人工智能（亚马逊A21)
Amazon Inspector
亚马逊Inspector
AWS Audit Manager
AWS审计管理器
正确答案 B
Amazon Augmented Al (Amazon A2l) enables you to build workflows for human review of machine learning predictions, allowing you to define and adjust confidence thresholds for when human intervention is required.
亚马逊增强型人工智能（AmazonA2I）使您能够构建用于人工审核机器学习预测的工作流，让您能够定义 和调整需要人工干预的置信度阈值。
A company wants to develop an Al assistant for employees to query internal data. Which AWS service will meet this requirement?
一家公司想要开发一款人工智能助手，供员工查询内部数据。 哪种AWS服务能满足这一要求？
Amazon Rekognition
亚马逊Rekognition
Amazon Textract
亚马逊Textract
Amazon Lex
亚马逊Lex
Amazon Q Business
亚马逊Q商业版
正确答案 D
Amazon Q Business is designed to build generative Al assistants for querying and interacting with internal organizational data, making it the ideal service for creating an Al assistant for employees to access company information.
亚马逊QBusiness旨在打造用于查询和交互内部组织数据的生成式人工智能助手，是为员工创建访问公司 信息的人工智能助手的理想服务。
A company wants to build and deploy ML models on AWS without writing any code. Which AWS service or feature meets these requirements?
一家公司希望在AWS上构建和部署机器学习模型，且无需编写任何代码。 哪个AWS服务或功能符合这些要求？
Amazon SageMaker Canvas
亚马逊 SageMaker Canvas
Amazon Rekognition
亚马逊Rekognition
AWS DeepRacer
AWS DeepRacer
Amazon Comprehend
亚马逊comprehend
正确答案 A
Amazon SageMaker Canvas allows users to build, train, and deploy machine learning models using a no- code visual interface, meeting the requirement to create ML solutions without writing any code.
亚马逊SageMakerCanvas使用户能够通过无代码的可视化界面构建、训练和部署机器学习模型，满足了无 需编写在荷代码即可创建机器学习解决方案的需求。合
| 已作答 | 0 |
| 答对 | 0 |
| 答错 | 0 |
| 未作答 | 0 |
| 用时 | — |