Semantic search boosts effectiveness by retrieving documents or data that closely match the user’s intent, even if the wording differs, thus providing the LLM with more relevant context for generation. Option A is unrelated —it describes machine translation. Option B refers to summarization, not retrieval. Option C describes voice interaction, which isn’t part of semantic search functionality. : Semantic search enables intent -aware retrieval, improving LLM accuracy and relevance in responses. When using the OCI Generative AI service playground, which of the following functionalities is NOT available?
APIs. How can you monitor the performance of a deployed model endpoint in OCI Generative AI?
crucial to optimize for accuracy?
D. To collect user feedback on the quality and creativity of the generated text. Explanation: Logging helps developers trace the execution flow, capture errors, and analyze how the LLM arrived at a certain output, which is essential for debugging and improving the application. Option A is incorrect —logging may actually add slight overhead. Option B is more related to auditing than execution tracing. Option D relates to evaluation, not the technical diagnostic process. : Logs provide visibility into LLM behavior, aiding troubleshooting and performance refinement. What are dedicated AI clusters used for in the context of OCI Generative AI?
buffers, not prompts. Option B may be part of the prompt but isn’t its core purpose. Option D relates to memory management, not prompt design. : Prompts shape the model’s response by clearly setting intent and scope at each stage in LangChain. When troubleshooting issues with a dedicated AI cluster for OCI Generative AI, where can you access logs and performance metrics for analysis?
When fine -tuning a pre -trained model in OCI Generative AI, what does the T -Few technique offer?
