100 Common AI terminologies for QA<\/title>\r\n <style>\r\n .terminoligy-wrap {\r\n font-family: 'Roboto Slab' !important;\r\n background: #fff;\r\n margin: 0;\r\n padding: 0;\r\n color: #333;\r\n }\r\n\r\n .terminoligy-wrap * {\r\n font-family: 'Roboto Slab' !important;\r\n }\r\n\r\n .terminoligy-wrap .container {\r\n max-width: 900px;\r\n margin: 0 auto;\r\n padding: 20px;\r\n }\r\n\r\n .terminoligy-wrap h1 {\r\n font-size: 2rem;\r\n margin-bottom: 20px;\r\n }\r\n\r\n .terminoligy-wrap .search-box {\r\n width: 100%;\r\n padding: 10px 15px;\r\n border: 1px solid #CCEEFF;\r\n border-radius: 8px !important;\r\n margin-bottom: 25px;\r\n font-size: 1rem;\r\n }\r\n\r\n .terminoligy-wrap input[type=text].search-box:focus {\r\n border-color: #e97713 !important;\r\n }\r\n\r\n .terminoligy-wrap .card {\r\n display: flex;\r\n align-items: flex-start;\r\n background: linear-gradient(to right, #F0FAFF, #FFF7F0) !important;\r\n border-radius: 12px;\r\n padding: 15px 20px;\r\n margin-bottom: 15px;\r\n border: 1px solid #CCEEFF;\r\n \/* box-shadow: 0 2px 6px rgba(0,0,0,0.1); *\/\r\n transition: transform 0.2s;\r\n }\r\n\r\n .terminoligy-wrap .card:hover {\r\n transform: translateY(-2px);\r\n }\r\n\r\n .terminoligy-wrap .icon {\r\n width: 40px;\r\n height: 40px;\r\n background: #CCEEFF;\r\n border-radius: 50%;\r\n display: flex;\r\n align-items: center;\r\n justify-content: center;\r\n font-weight: bold;\r\n margin-right: 15px;\r\n flex-shrink: 0;\r\n font-size: 1rem;\r\n }\r\n\r\n .terminoligy-wrap .card-content {\r\n flex: 1;\r\n }\r\n\r\n .terminoligy-wrap .term {\r\n font-weight: bold;\r\n font-size: 1.1rem;\r\n margin-bottom: 5px;\r\n }\r\n\r\n .terminoligy-wrap .definition {\r\n font-size: 0.95rem;\r\n line-height: 1.4;\r\n color: #555;\r\n }\r\n <\/style>\r\n<\/head>\r\n<body>\r\n <div class=\"container\">\r\n <h1>100 Thu\u1eadt ng\u1eef AI Th\u00f4ng d\u1ee5ng d\u00e0nh cho QA<\/h1>\r\n <input type=\"text\" class=\"search-box\" placeholder=\"T\u1eeb kh\u00f3a t\u00ecm ki\u1ebfm...\" onkeyup=\"filterCards()\">\r\n\r\n <div id=\"cards-container\">\r\n\r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Giao th\u1ee9c A2A<\/div>\r\n <div class=\"definition\">Ph\u01b0\u01a1ng ph\u00e1p giao ti\u1ebfp \u1ee8ng d\u1ee5ng sang \u1ee8ng d\u1ee5ng. QA c\u00f3 th\u1ec3 ki\u1ec3m tra \u0111\u1ec3 \u0111\u1ea3m b\u1ea3o c\u00e1c t\u01b0\u01a1ng t\u00e1c t\u1ef1 \u0111\u1ed9ng n\u00e0y an to\u00e0n v\u00e0 ho\u1ea1t \u0111\u1ed9ng \u0111\u00fang nh\u01b0 mong \u0111\u1ee3i.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Acceptance Thresholds (Metrics)<\/div>\r\n <div class=\"definition\">Minimum performance levels for release. The product team defines these threholds and QA would assess if the model is ready for deployment.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Accuracy (Metric)<\/div>\r\n <div class=\"definition\">Percentage of correct predictions. QA uses it for assessing overall model correctness on test sets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Adaptive Testing<\/div>\r\n <div class=\"definition\">Testing that changes as the AI evolves. QA continuously adjusts test scenarios to validate model evolving behavior.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Adversarial<\/div>\r\n <div class=\"definition\">Intentional input manipulation to mislead models. QA tests model robustness against such attacks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AGI<\/div>\r\n <div class=\"definition\">Artificial General Intelligence: A theoretical AI that can understand and perform any task like a human. It’s essential for understanding AI evolution.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Agents<\/div>\r\n <div class=\"definition\">Autonomous components that can make decisions and perform tasks without human intervention. QA may need to test for autonomy, reliability, and safety.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Alignment<\/div>\r\n <div class=\"definition\">Ensuring AI behaviors stay aligned with human goals and ethical expectations. QA often checks for safety, fairness, and unintended behaviors.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Model<\/div>\r\n <div class=\"definition\">A trained system (e.g., neural network) that maps inputs to outputs. Testers often validate it against expected performance, accuracy, and edge cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Wrapper<\/div>\r\n <div class=\"definition\">A software layer or application that simplifies access to AI models. QA would test like testing any software application in the context of AI-driven functionalities.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Algorithm<\/div>\r\n <div class=\"definition\">The underlying set of rules or logic that the AI follows. Testers may verify logical consistency and traceability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">ANN (Artificial Neural Network)<\/div>\r\n <div class=\"definition\">A machine learning model inspired by the brain. QA may test the outputs for accuracy, edge case handling, overfitting\/underfitting issues, etc.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Artificial Intelligence<\/div>\r\n <div class=\"definition\">Computer systems that mimic human intelligence to solve problems. QA testers evaluate how reliably and safely they perform tasks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Assistant Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">The AI\u2019s response behavior. QA tests for correctness, tone, and task-following accuracy.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Autonomous<\/div>\r\n <div class=\"definition\">Refers to systems that operate independently. QA tests to validate if agents predictably as expected and safely.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Bias<\/div>\r\n <div class=\"definition\">When a model shows unfair preferences. QA tests for performance across diverse groups to detect bias.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Bidirectional Testing<\/div>\r\n <div class=\"definition\">Evaluates how input affects output and how outputs may influence subsequent inputs or prompts. QA tests both directions to detect inconsistencies or circular dependencies.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Chain-of-Thought (CoT) Prompting<\/div>\r\n <div class=\"definition\">Encourages the AI to reason step-by-step. QA checks each reasoning step for correctness and coherence.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Chatbot<\/div>\r\n <div class=\"definition\">An AI system that interacts with users in natural language. Testers evaluate language understanding, edge cases, and intent coverage.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">CNN (Convolutional Neural Network)<\/div>\r\n <div class=\"definition\">A neural network best for image processing. QA tests image classification accuracy and robustness to distortions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Computer Vision<\/div>\r\n <div class=\"definition\">AI interpreting visual input. QA tests include edge detection, object recognition accuracy, and robustness to noise.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Concept Drift<\/div>\r\n <div class=\"definition\">This happens when what a model is predicting starts to mean something different over time\u2014like when \u201cfraud\u201d in financial data changes due to new scam tactics. QA testers check if the model\u2019s predictions still match current realities and whether labels need to be updated.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context<\/div>\r\n <div class=\"definition\">Relevant data history an AI uses for decision-making. Testers test for it\u2019s properly maintained and used.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context Injection Testing<\/div>\r\n <div class=\"definition\">Simulates scenarios where contextual inputs could manipulate or interfere with system prompts. QA verifies prompt isolation and defense layers.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context Window<\/div>\r\n <div class=\"definition\">The max input text an LLM can consider. QA checks truncation effects and response degradation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Continuous Learning<\/div>\r\n <div class=\"definition\">AI improves by learning from new data. QA tests for model updates while it does not introduce regressions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Co-piloting<\/div>\r\n <div class=\"definition\">AI-assisted development or testing support. QA validates suggestion accuracy, usefulness, and workflow integration.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Customize ChatGPT<\/div>\r\n <div class=\"definition\">Modifying ChatGPT\u2019s behavior using system instructions. It’s a useful tool to customize ChatGPT for better quality of the outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Dependency<\/div>\r\n <div class=\"definition\">AI relies heavily on training data. QA test for diverse, representative, and high-quality datasets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Drift<\/div>\r\n <div class=\"definition\">When incoming data shifts from training data. QA detects shifts that could lead to performance and accuracy drops.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Science<\/div>\r\n <div class=\"definition\">Field focused on extracting insights from data. QA may validate models built by data scientists and the correctness of data pipelines.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Dataset<\/div>\r\n <div class=\"definition\">A collection of data used for training or testing models. QA checks data quality, labeling accuracy, and bias.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Deep Learning<\/div>\r\n <div class=\"definition\">Subset of ML with layered neural networks. QA may test for generalization, overfitting, and inference correctness.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Embedding<\/div>\r\n <div class=\"definition\">Numeric vectors that represent words\/phrases. QA validates clustering, similarity scores, and semantic consistency.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ethics or Ethical Considerations<\/div>\r\n <div class=\"definition\">Responsible AI usage. QA tests for bias, misuse, and harmful outcomes.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Explainability<\/div>\r\n <div class=\"definition\">How well humans can understand AI decisions. QA checks that explanations are available, truthful, and useful.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">F1-Score (Metric)<\/div>\r\n <div class=\"definition\">Balance between precision and recall. QA uses it for performance in imbalanced datasets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Fairness<\/div>\r\n <div class=\"definition\">Equity in model outputs across users. QA tests for no group is disadvantaged unfairly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Few-shot Prompting<\/div>\r\n <div class=\"definition\">AI is given a few examples before answering. QA validates learning from small examples and matching output format.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Fine-tuning<\/div>\r\n <div class=\"definition\">Additional training on specific data. QA compares pre- and post-finetuned performance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Foundation Model<\/div>\r\n <div class=\"definition\">A large pre-trained model adaptable to many tasks. QA evaluates adaptability, efficiency, and risks of over-generalization.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Frequency Penalty<\/div>\r\n <div class=\"definition\">Reduces repetition by lowering likelihood of repeated words. QA evaluates its effect on redundancy and coherence.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">G.U.I.D.E. Prompting Formular<\/div>\r\n <div class=\"definition\">A structured way (Goals-Users-Instrucions-Details-Examples) to design effective prompts. QA uses it construct better prompts to elicit consistent and relevant AI outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Generative AI<\/div>\r\n <div class=\"definition\">AI that creates content (text, images, etc.). QA validates creativity vs. control, safety, and factuality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">GPU<\/div>\r\n <div class=\"definition\">Graphics hardware used to run AI efficiently. QA may monitor GPU utilization and performance under load.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ground Truth<\/div>\r\n <div class=\"definition\">The labeled data used as the correct answer during testing. Testers validate model outputs against ground truth.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Hallucination<\/div>\r\n <div class=\"definition\">When the model generates false or misleading content. Testers identify and report incorrect or nonsensical outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Human-in-the-Loop (HITL) Testing<\/div>\r\n <div class=\"definition\">Humans validate or correct model decisions. QA includes HITL as part of accuracy assurance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Inference<\/div>\r\n <div class=\"definition\">The model making predictions from inputs. QA test for predictions are accurate, fast, and explainable.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Instruction Collision Testing<\/div>\r\n <div class=\"definition\">Tests what happens when multiple instructions conflict. QA checks if the model prioritizes or blends conflicting directions responsibly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">LLM (Large Language Model)<\/div>\r\n <div class=\"definition\">A massive text-trained model like GPT-4. QA tests outputs for correctness, reliability, and adherence to constraints.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Machine Learning<\/div>\r\n <div class=\"definition\">AI that learns from data. QA tests ML systems across different datasets, distributions, and use cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Max Tokens<\/div>\r\n <div class=\"definition\">The maximum number of words or characters the AI can generate. QA testers validate that responses respect this limit and don\u2019t truncate unexpectedly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">MCP (Model Context Protocol)<\/div>\r\n <div class=\"definition\">A protocol framework for passing background information\u2014like user history, settings, or task data\u2014to AI models so they can respond more accurately. QA testers validate that this context is delivered correctly and that the model behaves appropriately based on it.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">MLOps<\/div>\r\n <div class=\"definition\">Operational practices for ML lifecycle. QA may test deployment stability, versioning, and monitoring processes.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Model Architecture<\/div>\r\n <div class=\"definition\">The design or structure of the AI model. QA may test performance differences across architectures.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Model Drift<\/div>\r\n <div class=\"definition\">When model behavior changes over time. QA monitors for accuracy degradation or unexpected predictions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Multimodal Testing<\/div>\r\n <div class=\"definition\">Validates models that use more than one input type (e.g., text + image). QA checks if the system processes and combines inputs correctly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Natural Language Processing (NLP)<\/div>\r\n <div class=\"definition\">AI that processes and understands human language. Testers validate intent recognition, language accuracy, and contextual responses.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Neural Network<\/div>\r\n <div class=\"definition\">A type of model inspired by the human brain. QA may test for learning is effective and interpretable where possible.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Overfitting<\/div>\r\n <div class=\"definition\">When the model performs well on training data but poorly on new data. QA uses holdout or test sets to detect it.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Parameters<\/div>\r\n <div class=\"definition\">The internal weights the model learns. QA doesn\u2019t directly test them, but they influence output quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Precision (Metric)<\/div>\r\n <div class=\"definition\">Correctness among positive predictions. QA uses it to check for false positives.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Presence Penalty<\/div>\r\n <div class=\"definition\">Discourages using words already mentioned. QA checks for novelty and logical flow in results.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Pre-trained Transformer<\/div>\r\n <div class=\"definition\">A model already trained on large data and adapted for new tasks. QA tests adaptation quality and residual knowledge.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Pretraining<\/div>\r\n <div class=\"definition\">Initial large-scale training on generic data. QA may test for knowledge transfer during fine-tuning.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ch\u00ednh s\u00e1ch<\/div>\r\n <div class=\"definition\">Protection of user data. QA checks for data leaks, logging issues, and compliance with data handling policies.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Probabilistic Outputs<\/div>\r\n <div class=\"definition\">AI provides outputs with varying confidence. QA evaluates certainty thresholds and result variation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Prompt Engineering<\/div>\r\n <div class=\"definition\">Crafting input prompts to guide model behavior. QA creates prompt suites to test consistency and compliance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde9<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Prompt Entanglement Testing<\/div>\r\n <div class=\"definition\">Examines whether earlier prompts unintentionally affect later responses. QA ensures prompt sessions are scoped correctly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">RAG (Retrieval-Augmented Generation)<\/div>\r\n <div class=\"definition\">A model architecture that uses search results to enhance output. QA validates grounding, retrieval accuracy, and source attribution.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Reasoning Model<\/div>\r\n <div class=\"definition\">Models designed for logical tasks. QA tests chain of thought, deduction accuracy, and reasoning robustness.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde9<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Recall (Metric)<\/div>\r\n <div class=\"definition\">Ability to find all relevant positives. QA uses it to check for false negatives.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Reinforcement Learning<\/div>\r\n <div class=\"definition\">Learning through reward signals. QA verifies stability, convergence, and policy safety.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Repeatability Testing<\/div>\r\n <div class=\"definition\">Checks whether a model consistently gives the same output when run multiple times under the same conditions. QA uses it to assess determinism and debugging reliability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Response Format<\/div>\r\n <div class=\"definition\">The structure or schema of the model’s output (e.g., JSON, markdown). QA verifies that formatting meets application needs and is parseable.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">RNN (Recurrent Neural Network)<\/div>\r\n <div class=\"definition\">A network for sequential data (e.g., text, time series). QA checks for memory of past inputs and proper sequence predictions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Robustness<\/div>\r\n <div class=\"definition\">Stability of model outputs under noisy, diverse, or adversarial inputs. QA stress-tests edge conditions and rare cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Semi-supervised Learning (Algorithm)<\/div>\r\n <div class=\"definition\">A method using both labeled and unlabeled data. QA may tests for generalization and monitors performance gaps.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Sensitivity Analysis<\/div>\r\n <div class=\"definition\">Analyzes how small changes in input affect output. QA uses it to identify fragile or overly sensitive behaviors.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Stop Sequences<\/div>\r\n <div class=\"definition\">Text that tells the AI where to stop generating. QA verifies proper cutoffs to avoid extra or incomplete responses.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Supervised Learning<\/div>\r\n <div class=\"definition\">Training with labeled data. QA uses test sets and metrics (accuracy, F1) to validate.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">System Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">Defines the AI\u2019s behavior context (e.g., expert, assistant). QA test for role consistency and output alignment.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Temperature<\/div>\r\n <div class=\"definition\">Affects randomness\u2014higher values produce more creative, varied responses. QA tunes this for output consistency vs creativity.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Text-to-Image Generation<\/div>\r\n <div class=\"definition\">Creating images from text prompts. QA assesses image quality, prompt alignment, and potential harm.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Tokenization<\/div>\r\n <div class=\"definition\">Splitting text into tokens for processing. QA test for token limits and structure aren\u2019t broken.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Top P<\/div>\r\n <div class=\"definition\">Controls randomness by limiting tokens to a cumulative probability. QA tests how this affects output diversity and determinism.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">TPU<\/div>\r\n <div class=\"definition\">Tensor Processing Unit \u2013 Google\u2019s AI chip.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Training<\/div>\r\n <div class=\"definition\">The full process of model learning. QA checks for learning progress, convergence, and generalization.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Training Data<\/div>\r\n <div class=\"definition\">The dataset used to teach the model. QA assesses quality, diversity, and labeling accuracy.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Transformer<\/div>\r\n <div class=\"definition\">Model design used in most LLMs. QA focuses on attention behaviors, context limits, and generation stability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Transformer Model<\/div>\r\n <div class=\"definition\">A model using attention mechanisms. QA test for handling of long sequences and correct interpretation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Tree-of-Thought (ToT) Prompting<\/div>\r\n <div class=\"definition\">AI explores multiple reasoning paths like decision trees. QA test for completeness and evaluates path quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Unsupervised Learning<\/div>\r\n <div class=\"definition\">Learning without labeled data. QA checks discovered patterns and use in downstream tasks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">User Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">The prompt or question from the human user. QA designs diverse prompts to validate AI understanding.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Variance\/Configuration Testing<\/div>\r\n <div class=\"definition\">Tests how changes in settings (e.g., temperature, top_p) affect model behavior. QA evaluates performance and stability across configurations.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Vibe Coding<\/div>\r\n <div class=\"definition\">Natural language-driven coding tools. QA tests IDE integration, code accuracy, and developer prompts.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Weights<\/div>\r\n <div class=\"definition\">Numerical values the model adjusts during training. QA indirectly assesses them via output quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Zero-Shot Learning<\/div>\r\n <div class=\"definition\">Making predictions without task-specific training. QA assesses generalization and failure cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Zero-shot Prompting<\/div>\r\n <div class=\"definition\">AI performs a task without specific examples. QA evaluates if it handles generalization effectively.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <\/div>\r\n <\/div>\r\n\r\n <script>\r\n function filterCards() {\r\n const input = document.querySelector('.search-box').value.toLowerCase();\r\n const cards = document.querySelectorAll('.card');\r\n cards.forEach(card => {\r\n const term = card.querySelector('.term').innerText.toLowerCase();\r\n const definition = card.querySelector('.definition').innerText.toLowerCase();\r\n if (term.includes(input) || definition.includes(input)) {\r\n card.style.display = 'flex';\r\n } else {\r\n card.style.display = 'none';\r\n }\r\n });\r\n }\r\n <\/script>\r\n<\/body>\r\n<\/html>\n<\/div><\/div>","protected":false},"excerpt":{"rendered":"<p>100 Common AI terminologies for QA 100 Common AI terminologies for QA \ud83e\udde0 A2A Protocol Application-to-Application communication method. QA may test for these automated interactions are secure and behave as expected. \ud83d\udce6 Acceptance Thresholds (Metrics) Minimum performance levels for release. … <\/p>","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_coblocks_attr":"","_coblocks_dimensions":"","_coblocks_responsive_height":"","_coblocks_accordion_ie_support":"","footnotes":""},"class_list":["post-23665","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/comments?post=23665"}],"version-history":[{"count":2,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665\/revisions"}],"predecessor-version":[{"id":23667,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665\/revisions\/23667"}],"wp:attachment":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/media?parent=23665"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}

\n\r\n\r\n\r\n \r\n \r\n 100 Common AI terminologies for QA<\/title>\r\n <style>\r\n .terminoligy-wrap {\r\n font-family: 'Roboto Slab' !important;\r\n background: #fff;\r\n margin: 0;\r\n padding: 0;\r\n color: #333;\r\n }\r\n\r\n .terminoligy-wrap * {\r\n font-family: 'Roboto Slab' !important;\r\n }\r\n\r\n .terminoligy-wrap .container {\r\n max-width: 900px;\r\n margin: 0 auto;\r\n padding: 20px;\r\n }\r\n\r\n .terminoligy-wrap h1 {\r\n font-size: 2rem;\r\n margin-bottom: 20px;\r\n }\r\n\r\n .terminoligy-wrap .search-box {\r\n width: 100%;\r\n padding: 10px 15px;\r\n border: 1px solid #CCEEFF;\r\n border-radius: 8px !important;\r\n margin-bottom: 25px;\r\n font-size: 1rem;\r\n }\r\n\r\n .terminoligy-wrap input[type=text].search-box:focus {\r\n border-color: #e97713 !important;\r\n }\r\n\r\n .terminoligy-wrap .card {\r\n display: flex;\r\n align-items: flex-start;\r\n background: linear-gradient(to right, #F0FAFF, #FFF7F0) !important;\r\n border-radius: 12px;\r\n padding: 15px 20px;\r\n margin-bottom: 15px;\r\n border: 1px solid #CCEEFF;\r\n \/* box-shadow: 0 2px 6px rgba(0,0,0,0.1); *\/\r\n transition: transform 0.2s;\r\n }\r\n\r\n .terminoligy-wrap .card:hover {\r\n transform: translateY(-2px);\r\n }\r\n\r\n .terminoligy-wrap .icon {\r\n width: 40px;\r\n height: 40px;\r\n background: #CCEEFF;\r\n border-radius: 50%;\r\n display: flex;\r\n align-items: center;\r\n justify-content: center;\r\n font-weight: bold;\r\n margin-right: 15px;\r\n flex-shrink: 0;\r\n font-size: 1rem;\r\n }\r\n\r\n .terminoligy-wrap .card-content {\r\n flex: 1;\r\n }\r\n\r\n .terminoligy-wrap .term {\r\n font-weight: bold;\r\n font-size: 1.1rem;\r\n margin-bottom: 5px;\r\n }\r\n\r\n .terminoligy-wrap .definition {\r\n font-size: 0.95rem;\r\n line-height: 1.4;\r\n color: #555;\r\n }\r\n <\/style>\r\n<\/head>\r\n<body>\r\n <div class=\"container\">\r\n <h1>100 Thu\u1eadt ng\u1eef AI Th\u00f4ng d\u1ee5ng d\u00e0nh cho QA<\/h1>\r\n <input type=\"text\" class=\"search-box\" placeholder=\"T\u1eeb kh\u00f3a t\u00ecm ki\u1ebfm...\" onkeyup=\"filterCards()\">\r\n\r\n <div id=\"cards-container\">\r\n\r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Giao th\u1ee9c A2A<\/div>\r\n <div class=\"definition\">Ph\u01b0\u01a1ng ph\u00e1p giao ti\u1ebfp \u1ee8ng d\u1ee5ng sang \u1ee8ng d\u1ee5ng. QA c\u00f3 th\u1ec3 ki\u1ec3m tra \u0111\u1ec3 \u0111\u1ea3m b\u1ea3o c\u00e1c t\u01b0\u01a1ng t\u00e1c t\u1ef1 \u0111\u1ed9ng n\u00e0y an to\u00e0n v\u00e0 ho\u1ea1t \u0111\u1ed9ng \u0111\u00fang nh\u01b0 mong \u0111\u1ee3i.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Acceptance Thresholds (Metrics)<\/div>\r\n <div class=\"definition\">Minimum performance levels for release. The product team defines these threholds and QA would assess if the model is ready for deployment.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Accuracy (Metric)<\/div>\r\n <div class=\"definition\">Percentage of correct predictions. QA uses it for assessing overall model correctness on test sets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Adaptive Testing<\/div>\r\n <div class=\"definition\">Testing that changes as the AI evolves. QA continuously adjusts test scenarios to validate model evolving behavior.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Adversarial<\/div>\r\n <div class=\"definition\">Intentional input manipulation to mislead models. QA tests model robustness against such attacks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AGI<\/div>\r\n <div class=\"definition\">Artificial General Intelligence: A theoretical AI that can understand and perform any task like a human. It’s essential for understanding AI evolution.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Agents<\/div>\r\n <div class=\"definition\">Autonomous components that can make decisions and perform tasks without human intervention. QA may need to test for autonomy, reliability, and safety.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Alignment<\/div>\r\n <div class=\"definition\">Ensuring AI behaviors stay aligned with human goals and ethical expectations. QA often checks for safety, fairness, and unintended behaviors.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Model<\/div>\r\n <div class=\"definition\">A trained system (e.g., neural network) that maps inputs to outputs. Testers often validate it against expected performance, accuracy, and edge cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">AI Wrapper<\/div>\r\n <div class=\"definition\">A software layer or application that simplifies access to AI models. QA would test like testing any software application in the context of AI-driven functionalities.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Algorithm<\/div>\r\n <div class=\"definition\">The underlying set of rules or logic that the AI follows. Testers may verify logical consistency and traceability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">ANN (Artificial Neural Network)<\/div>\r\n <div class=\"definition\">A machine learning model inspired by the brain. QA may test the outputs for accuracy, edge case handling, overfitting\/underfitting issues, etc.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Artificial Intelligence<\/div>\r\n <div class=\"definition\">Computer systems that mimic human intelligence to solve problems. QA testers evaluate how reliably and safely they perform tasks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Assistant Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">The AI\u2019s response behavior. QA tests for correctness, tone, and task-following accuracy.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Autonomous<\/div>\r\n <div class=\"definition\">Refers to systems that operate independently. QA tests to validate if agents predictably as expected and safely.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcdd<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Bias<\/div>\r\n <div class=\"definition\">When a model shows unfair preferences. QA tests for performance across diverse groups to detect bias.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Bidirectional Testing<\/div>\r\n <div class=\"definition\">Evaluates how input affects output and how outputs may influence subsequent inputs or prompts. QA tests both directions to detect inconsistencies or circular dependencies.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Chain-of-Thought (CoT) Prompting<\/div>\r\n <div class=\"definition\">Encourages the AI to reason step-by-step. QA checks each reasoning step for correctness and coherence.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Chatbot<\/div>\r\n <div class=\"definition\">An AI system that interacts with users in natural language. Testers evaluate language understanding, edge cases, and intent coverage.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">CNN (Convolutional Neural Network)<\/div>\r\n <div class=\"definition\">A neural network best for image processing. QA tests image classification accuracy and robustness to distortions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Computer Vision<\/div>\r\n <div class=\"definition\">AI interpreting visual input. QA tests include edge detection, object recognition accuracy, and robustness to noise.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Concept Drift<\/div>\r\n <div class=\"definition\">This happens when what a model is predicting starts to mean something different over time\u2014like when \u201cfraud\u201d in financial data changes due to new scam tactics. QA testers check if the model\u2019s predictions still match current realities and whether labels need to be updated.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context<\/div>\r\n <div class=\"definition\">Relevant data history an AI uses for decision-making. Testers test for it\u2019s properly maintained and used.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context Injection Testing<\/div>\r\n <div class=\"definition\">Simulates scenarios where contextual inputs could manipulate or interfere with system prompts. QA verifies prompt isolation and defense layers.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Context Window<\/div>\r\n <div class=\"definition\">The max input text an LLM can consider. QA checks truncation effects and response degradation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Continuous Learning<\/div>\r\n <div class=\"definition\">AI improves by learning from new data. QA tests for model updates while it does not introduce regressions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Co-piloting<\/div>\r\n <div class=\"definition\">AI-assisted development or testing support. QA validates suggestion accuracy, usefulness, and workflow integration.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Customize ChatGPT<\/div>\r\n <div class=\"definition\">Modifying ChatGPT\u2019s behavior using system instructions. It’s a useful tool to customize ChatGPT for better quality of the outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Dependency<\/div>\r\n <div class=\"definition\">AI relies heavily on training data. QA test for diverse, representative, and high-quality datasets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Drift<\/div>\r\n <div class=\"definition\">When incoming data shifts from training data. QA detects shifts that could lead to performance and accuracy drops.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Data Science<\/div>\r\n <div class=\"definition\">Field focused on extracting insights from data. QA may validate models built by data scientists and the correctness of data pipelines.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Dataset<\/div>\r\n <div class=\"definition\">A collection of data used for training or testing models. QA checks data quality, labeling accuracy, and bias.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Deep Learning<\/div>\r\n <div class=\"definition\">Subset of ML with layered neural networks. QA may test for generalization, overfitting, and inference correctness.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Embedding<\/div>\r\n <div class=\"definition\">Numeric vectors that represent words\/phrases. QA validates clustering, similarity scores, and semantic consistency.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ethics or Ethical Considerations<\/div>\r\n <div class=\"definition\">Responsible AI usage. QA tests for bias, misuse, and harmful outcomes.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Explainability<\/div>\r\n <div class=\"definition\">How well humans can understand AI decisions. QA checks that explanations are available, truthful, and useful.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">F1-Score (Metric)<\/div>\r\n <div class=\"definition\">Balance between precision and recall. QA uses it for performance in imbalanced datasets.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Fairness<\/div>\r\n <div class=\"definition\">Equity in model outputs across users. QA tests for no group is disadvantaged unfairly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Few-shot Prompting<\/div>\r\n <div class=\"definition\">AI is given a few examples before answering. QA validates learning from small examples and matching output format.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Fine-tuning<\/div>\r\n <div class=\"definition\">Additional training on specific data. QA compares pre- and post-finetuned performance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Foundation Model<\/div>\r\n <div class=\"definition\">A large pre-trained model adaptable to many tasks. QA evaluates adaptability, efficiency, and risks of over-generalization.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Frequency Penalty<\/div>\r\n <div class=\"definition\">Reduces repetition by lowering likelihood of repeated words. QA evaluates its effect on redundancy and coherence.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">G.U.I.D.E. Prompting Formular<\/div>\r\n <div class=\"definition\">A structured way (Goals-Users-Instrucions-Details-Examples) to design effective prompts. QA uses it construct better prompts to elicit consistent and relevant AI outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Generative AI<\/div>\r\n <div class=\"definition\">AI that creates content (text, images, etc.). QA validates creativity vs. control, safety, and factuality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">GPU<\/div>\r\n <div class=\"definition\">Graphics hardware used to run AI efficiently. QA may monitor GPU utilization and performance under load.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ground Truth<\/div>\r\n <div class=\"definition\">The labeled data used as the correct answer during testing. Testers validate model outputs against ground truth.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Hallucination<\/div>\r\n <div class=\"definition\">When the model generates false or misleading content. Testers identify and report incorrect or nonsensical outputs.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Human-in-the-Loop (HITL) Testing<\/div>\r\n <div class=\"definition\">Humans validate or correct model decisions. QA includes HITL as part of accuracy assurance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Inference<\/div>\r\n <div class=\"definition\">The model making predictions from inputs. QA test for predictions are accurate, fast, and explainable.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Instruction Collision Testing<\/div>\r\n <div class=\"definition\">Tests what happens when multiple instructions conflict. QA checks if the model prioritizes or blends conflicting directions responsibly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">LLM (Large Language Model)<\/div>\r\n <div class=\"definition\">A massive text-trained model like GPT-4. QA tests outputs for correctness, reliability, and adherence to constraints.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Machine Learning<\/div>\r\n <div class=\"definition\">AI that learns from data. QA tests ML systems across different datasets, distributions, and use cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Max Tokens<\/div>\r\n <div class=\"definition\">The maximum number of words or characters the AI can generate. QA testers validate that responses respect this limit and don\u2019t truncate unexpectedly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">MCP (Model Context Protocol)<\/div>\r\n <div class=\"definition\">A protocol framework for passing background information\u2014like user history, settings, or task data\u2014to AI models so they can respond more accurately. QA testers validate that this context is delivered correctly and that the model behaves appropriately based on it.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">MLOps<\/div>\r\n <div class=\"definition\">Operational practices for ML lifecycle. QA may test deployment stability, versioning, and monitoring processes.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Model Architecture<\/div>\r\n <div class=\"definition\">The design or structure of the AI model. QA may test performance differences across architectures.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Model Drift<\/div>\r\n <div class=\"definition\">When model behavior changes over time. QA monitors for accuracy degradation or unexpected predictions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Multimodal Testing<\/div>\r\n <div class=\"definition\">Validates models that use more than one input type (e.g., text + image). QA checks if the system processes and combines inputs correctly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Natural Language Processing (NLP)<\/div>\r\n <div class=\"definition\">AI that processes and understands human language. Testers validate intent recognition, language accuracy, and contextual responses.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Neural Network<\/div>\r\n <div class=\"definition\">A type of model inspired by the human brain. QA may test for learning is effective and interpretable where possible.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Overfitting<\/div>\r\n <div class=\"definition\">When the model performs well on training data but poorly on new data. QA uses holdout or test sets to detect it.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Parameters<\/div>\r\n <div class=\"definition\">The internal weights the model learns. QA doesn\u2019t directly test them, but they influence output quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Precision (Metric)<\/div>\r\n <div class=\"definition\">Correctness among positive predictions. QA uses it to check for false positives.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Presence Penalty<\/div>\r\n <div class=\"definition\">Discourages using words already mentioned. QA checks for novelty and logical flow in results.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcca<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Pre-trained Transformer<\/div>\r\n <div class=\"definition\">A model already trained on large data and adapted for new tasks. QA tests adaptation quality and residual knowledge.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Pretraining<\/div>\r\n <div class=\"definition\">Initial large-scale training on generic data. QA may test for knowledge transfer during fine-tuning.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Ch\u00ednh s\u00e1ch<\/div>\r\n <div class=\"definition\">Protection of user data. QA checks for data leaks, logging issues, and compliance with data handling policies.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Probabilistic Outputs<\/div>\r\n <div class=\"definition\">AI provides outputs with varying confidence. QA evaluates certainty thresholds and result variation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Prompt Engineering<\/div>\r\n <div class=\"definition\">Crafting input prompts to guide model behavior. QA creates prompt suites to test consistency and compliance.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde9<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Prompt Entanglement Testing<\/div>\r\n <div class=\"definition\">Examines whether earlier prompts unintentionally affect later responses. QA ensures prompt sessions are scoped correctly.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">RAG (Retrieval-Augmented Generation)<\/div>\r\n <div class=\"definition\">A model architecture that uses search results to enhance output. QA validates grounding, retrieval accuracy, and source attribution.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Reasoning Model<\/div>\r\n <div class=\"definition\">Models designed for logical tasks. QA tests chain of thought, deduction accuracy, and reasoning robustness.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde9<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Recall (Metric)<\/div>\r\n <div class=\"definition\">Ability to find all relevant positives. QA uses it to check for false negatives.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udde0<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Reinforcement Learning<\/div>\r\n <div class=\"definition\">Learning through reward signals. QA verifies stability, convergence, and policy safety.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Repeatability Testing<\/div>\r\n <div class=\"definition\">Checks whether a model consistently gives the same output when run multiple times under the same conditions. QA uses it to assess determinism and debugging reliability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\ude84<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Response Format<\/div>\r\n <div class=\"definition\">The structure or schema of the model’s output (e.g., JSON, markdown). QA verifies that formatting meets application needs and is parseable.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udee0\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">RNN (Recurrent Neural Network)<\/div>\r\n <div class=\"definition\">A network for sequential data (e.g., text, time series). QA checks for memory of past inputs and proper sequence predictions.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Robustness<\/div>\r\n <div class=\"definition\">Stability of model outputs under noisy, diverse, or adversarial inputs. QA stress-tests edge conditions and rare cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Semi-supervised Learning (Algorithm)<\/div>\r\n <div class=\"definition\">A method using both labeled and unlabeled data. QA may tests for generalization and monitors performance gaps.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\uddbc\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Sensitivity Analysis<\/div>\r\n <div class=\"definition\">Analyzes how small changes in input affect output. QA uses it to identify fragile or overly sensitive behaviors.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Stop Sequences<\/div>\r\n <div class=\"definition\">Text that tells the AI where to stop generating. QA verifies proper cutoffs to avoid extra or incomplete responses.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Supervised Learning<\/div>\r\n <div class=\"definition\">Training with labeled data. QA uses test sets and metrics (accuracy, F1) to validate.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\ude80<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">System Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">Defines the AI\u2019s behavior context (e.g., expert, assistant). QA test for role consistency and output alignment.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Temperature<\/div>\r\n <div class=\"definition\">Affects randomness\u2014higher values produce more creative, varied responses. QA tunes this for output consistency vs creativity.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udfaf<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Text-to-Image Generation<\/div>\r\n <div class=\"definition\">Creating images from text prompts. QA assesses image quality, prompt alignment, and potential harm.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Tokenization<\/div>\r\n <div class=\"definition\">Splitting text into tokens for processing. QA test for token limits and structure aren\u2019t broken.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Top P<\/div>\r\n <div class=\"definition\">Controls randomness by limiting tokens to a cumulative probability. QA tests how this affects output diversity and determinism.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">TPU<\/div>\r\n <div class=\"definition\">Tensor Processing Unit \u2013 Google\u2019s AI chip.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udce6<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Training<\/div>\r\n <div class=\"definition\">The full process of model learning. QA checks for learning progress, convergence, and generalization.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83e\udd16<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Training Data<\/div>\r\n <div class=\"definition\">The dataset used to teach the model. QA assesses quality, diversity, and labeling accuracy.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Transformer<\/div>\r\n <div class=\"definition\">Model design used in most LLMs. QA focuses on attention behaviors, context limits, and generation stability.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\u2699\ufe0f<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Transformer Model<\/div>\r\n <div class=\"definition\">A model using attention mechanisms. QA test for handling of long sequences and correct interpretation.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Tree-of-Thought (ToT) Prompting<\/div>\r\n <div class=\"definition\">AI explores multiple reasoning paths like decision trees. QA test for completeness and evaluates path quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Unsupervised Learning<\/div>\r\n <div class=\"definition\">Learning without labeled data. QA checks discovered patterns and use in downstream tasks.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83c\udf10<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">User Role (In Prompt Engineering)<\/div>\r\n <div class=\"definition\">The prompt or question from the human user. QA designs diverse prompts to validate AI understanding.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd12<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Variance\/Configuration Testing<\/div>\r\n <div class=\"definition\">Tests how changes in settings (e.g., temperature, top_p) affect model behavior. QA evaluates performance and stability across configurations.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udd0d<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Vibe Coding<\/div>\r\n <div class=\"definition\">Natural language-driven coding tools. QA tests IDE integration, code accuracy, and developer prompts.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcc8<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Weights<\/div>\r\n <div class=\"definition\">Numerical values the model adjusts during training. QA indirectly assesses them via output quality.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udca1<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Zero-Shot Learning<\/div>\r\n <div class=\"definition\">Making predictions without task-specific training. QA assesses generalization and failure cases.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <div class=\"card\">\r\n <div class=\"icon\">\ud83d\udcda<\/div>\r\n <div class=\"card-content\">\r\n <div class=\"term\">Zero-shot Prompting<\/div>\r\n <div class=\"definition\">AI performs a task without specific examples. QA evaluates if it handles generalization effectively.<\/div>\r\n <\/div>\r\n <\/div>\r\n \r\n <\/div>\r\n <\/div>\r\n\r\n <script>\r\n function filterCards() {\r\n const input = document.querySelector('.search-box').value.toLowerCase();\r\n const cards = document.querySelectorAll('.card');\r\n cards.forEach(card => {\r\n const term = card.querySelector('.term').innerText.toLowerCase();\r\n const definition = card.querySelector('.definition').innerText.toLowerCase();\r\n if (term.includes(input) || definition.includes(input)) {\r\n card.style.display = 'flex';\r\n } else {\r\n card.style.display = 'none';\r\n }\r\n });\r\n }\r\n <\/script>\r\n<\/body>\r\n<\/html>\n<\/div><\/div>","protected":false},"excerpt":{"rendered":"<p>100 Common AI terminologies for QA 100 Common AI terminologies for QA \ud83e\udde0 A2A Protocol Application-to-Application communication method. QA may test for these automated interactions are secure and behave as expected. \ud83d\udce6 Acceptance Thresholds (Metrics) Minimum performance levels for release. … <\/p>","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_coblocks_attr":"","_coblocks_dimensions":"","_coblocks_responsive_height":"","_coblocks_accordion_ie_support":"","footnotes":""},"class_list":["post-23665","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/comments?post=23665"}],"version-history":[{"count":2,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665\/revisions"}],"predecessor-version":[{"id":23667,"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/pages\/23665\/revisions\/23667"}],"wp:attachment":[{"href":"https:\/\/shadeofhue.com\/vi\/wp-json\/wp\/v2\/media?parent=23665"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}