Practice: Introduction to Generative AI for Software Testing

Which branch of AI uses neural networks to automatically learn features from large and complex datasets?

Which statement BEST explains why an LLM may generate a plausible but incorrect answer?

A tester inputs a very long requirements document, but the LLM ignores details from the beginning. What is the MOST likely reason?

Which statement BEST distinguishes tokenization from embeddings?

Which TWO statements about LLMs are CORRECT? Statement A = They always produce the same output for the same prompt. Statement B = They require manual feature selection for each testing task. Statement C = They use numerical representations of language. Statement D = They may produce varied outputs due to probabilistic inference. Statement E = They cannot process test documentation.

A team needs a model for multi-step defect trend analysis and test effort prioritization. Which type is MOST suitable?

What is the defining characteristic of an instruction-tuned LLM?

Which statement about foundation LLMs is CORRECT?

Which statement BEST describes the relation between multimodal LLMs and vision-language models?

A tester provides a GUI screenshot and a user story to identify inconsistencies. Which capability is being used?

How are images typically prepared for processing in multimodal transformer models?

Which option is the BEST example of LLM support for requirements analysis?

Which TWO are key LLM capabilities for software testing? Capability A = Guaranteeing defect-free releases. Capability B = Eliminating the need for testers. Capability C = Generating test data with boundary values. Capability D = Summarizing test results and classifying anomalies. Capability E = Removing the need for requirements.

Which statement BEST describes LLM support for test oracle generation?

Which statement BEST differentiates AI chatbots from LLM-powered testing applications?

A tester wants quick clarification of possible test ideas during exploratory testing. Which interaction model is MOST suitable?

An organization wants automated defect report summarization inside its existing test management tool. Which approach BEST fits?