1Z0-1127-25 Hard Questions — 25 Difficult Questions

Question 1

Which THREE techniques effectively reduce query latency in a RAG system?

Accepted Answer

Pre-compute embeddings for all documents. Pre-computing embeddings for all documents eliminates the need to generate embeddings at query time, which is a computationally expensive step. By storing pre-computed vector representations, the system can directly perform similarity searches against the index, significantly reducing latency.

Answer

Use a larger generation model

Answer

Increase the number of shards

Question 2

A prompt engineer notices that the model sometimes generates outputs that include parts of the system prompt or user message verbatim. This is likely a symptom of which common prompt failure?

Accepted Answer

Prompt injection vulnerabilities. Prompt injection vulnerabilities can cause the model to leak or repeat the prompt itself. This is a known failure mode where the model confuses the input with output.

Answer

Ambiguous instructions

Answer

Conflicting requirements

Answer

Insufficient context

Question 3

During fine-tuning of a large language model on OCI, you notice that the model's performance on the validation set is not improving after several epochs, but the training loss continues to decrease. What is the most likely cause?

Accepted Answer

The model is overfitting to the training data.. When training loss decreases but validation performance stagnates or worsens, the model is overfitting to the training data. It memorizes the training examples but fails to generalize. A high learning rate might cause divergence, not this pattern. Too small training data can contribute to overfitting but is not the direct symptom. An unrepresentative validation set could cause mismatch, but the described pattern is classic overfitting.

Answer

The learning rate is too high.

Answer

The validation set is not representative.

Answer

The training data is too small.

Question 4

Which three characteristics of LLMs can lead to hallucinations? (Select THREE)

Accepted Answer

Ability to generate plausible-sounding text. LLMs are trained to generate text that is statistically plausible and coherent, but they lack mechanisms to verify factual accuracy. This means they can produce sentences that sound convincing and grammatically correct while being entirely false, which is a direct cause of hallucinations.

Answer

Overconfidence in predictions

Answer

Large vocabulary size

Question 5

Which three statements about transformer architecture are correct? (Choose three.)

Accepted Answer

The softmax function is used in the attention mechanism to normalize attention scores.. The softmax function is applied to the raw attention scores (the dot products between queries and keys) to convert them into a probability distribution that sums to 1. This normalization allows the model to assign a relative weight to each token in the sequence, ensuring that the weighted sum of values is stable and interpretable.

Answer

The feed-forward network applies a different set of weights for each token position.

Answer

The encoder-decoder structure is used in GPT models.

Question 6

An application mixes RAG with other data sources. The vector search returns too many irrelevant chunks. What is the best approach to filter them?

Accepted Answer

Use a reranker model. A reranker model (Option A) is the best approach because it takes the initial set of retrieved chunks and re-orders them based on semantic relevance to the query, effectively filtering out irrelevant chunks. Unlike simple vector similarity, a reranker uses cross-encoding to evaluate the query-chunk pair as a whole, which significantly improves precision when mixing RAG with other data sources.

Answer

Use exact search instead of ANN

Answer

Reduce the number of retrieved chunks

Answer

Increase chunk size

Question 7

A company runs batch inference jobs daily using the OCI Generative AI service. The current cost is higher than expected. Which change would most effectively reduce cost while maintaining throughput?

Accepted Answer

Switch from on-demand to dedicated AI cluster with batch endpoint.. Switching from on-demand to a dedicated AI cluster with a batch endpoint reduces cost because dedicated clusters provide reserved capacity at a lower per-token rate compared to on-demand pay-per-token pricing, and batch endpoints allow you to process multiple inference requests in a single job, amortizing overhead and reducing idle time. This combination directly addresses the high cost of per-request on-demand pricing while maintaining the same throughput for daily batch jobs.

Answer

Reduce the max token limit for all requests.

Answer

Use a larger model to reduce retries.

Answer

Increase the number of parallel requests to improve efficiency.

Question 8

Which THREE techniques are commonly used to improve the quality of text generation?

Accepted Answer

Temperature scaling. Temperature scaling is correct because it controls the randomness of token probability distributions by dividing logits before softmax; lower temperatures (e.g., 0.1) make the model more deterministic, while higher temperatures (e.g., 1.5) increase diversity. This directly influences the quality of generated text by balancing coherence and creativity.

Answer

Greedy decoding

Answer

Random sampling

Question 9

A team has deployed a generative AI model using OCI Data Science model deployment. The endpoint is behind a load balancer. Users report that after 5 minutes of inactivity, the first request takes over 30 seconds to respond, while subsequent requests are fast. What is the most likely cause and solution?

Accepted Answer

The model deployment has an idle timeout that scales down to zero; configure a minimum number of instances or use a warm-up request. The described behavior—first request after 5 minutes of inactivity taking over 30 seconds, with subsequent requests fast—is a classic symptom of an idle timeout that scales the model deployment to zero instances. OCI Data Science model deployments support auto-scaling with an idle timeout (default 5 minutes) that can reduce the number of instances to zero when no requests are received. When a new request arrives, it must wait for a new instance to spin up, causing the delay. The solution is to configure a minimum number of instances (e.g., 1) to keep the model warm, or use a warm-up request to prevent the idle timeout from triggering.

Answer

The load balancer is scaling based on CPU utilization; increase the CPU threshold

Answer

The VCN has a network latency issue; use a different availability domain

Answer

The inference code has a lazy initialization; pre-load the model in the deployment script

Question 10

A company deploys a large language model on a dedicated AI cluster with 4 nodes. The model requires 128 GB of memory per instance, but the nodes have only 64 GB each. During inference, the nodes experience out-of-memory errors. What is the best solution?

Accepted Answer

Enable model parallelism across nodes. Model parallelism splits the model's layers or parameters across multiple nodes, allowing the 128 GB model to be distributed across the 4 nodes (each with 64 GB) so that no single node exceeds its memory capacity. This is the best solution because it directly addresses the memory constraint without requiring additional hardware or sacrificing inference throughput, and it is a standard technique for deploying large language models on distributed AI clusters.

Answer

Increase the number of nodes to 8

Answer

Upgrade to higher memory node shapes

Answer

Reduce the batch size in inference requests

Question 11

Refer to the exhibit. A developer encounters this error. Which action should they take to resolve the issue?

Accepted Answer

Wait and retry after some time.. The error indicates a rate limit or throttling issue, typically returned by the OCI Generative AI service when the API request quota is exceeded. Waiting and retrying after the cooldown period allows the rate limit to reset, which is the correct resolution for transient throttling errors.

Answer

Change the model to cohere.command-light.

Answer

Increase the max-tokens value.

Answer

Decrease the temperature to 0.0.

Question 12

Which THREE components are essential for a production-grade generative AI deployment on OCI? (Select THREE)

Accepted Answer

OCI Logging for audit. A is correct because OCI Logging provides centralized audit logging for all API calls and resource changes in the generative AI deployment. This is essential for compliance, security monitoring, and troubleshooting in a production environment, as it captures detailed logs of model invocations, data access, and configuration changes.

Answer

OCI Vault for secrets

Answer

OCI Data Flow for data processing

Question 13

During deployment of a generative AI model, the inference endpoint returns high latency and timeouts. The model is deployed on a dedicated AI cluster with multiple nodes. What is the most likely cause?

Accepted Answer

The cluster nodes are configured with insufficient parallelism or the model is not properly parallelized across nodes. High latency and timeouts in a distributed AI inference deployment typically indicate that the model workload is not efficiently distributed across the cluster nodes. Option C is correct because insufficient parallelism—either due to misconfigured node resources (e.g., insufficient vCPUs, GPU cores, or memory bandwidth) or improper model sharding/parallelization—causes some nodes to become bottlenecks while others remain underutilized, leading to queuing delays and eventual timeouts.

Answer

The inference request batch size is too small

Answer

The model is too large for the cluster memory

Answer

The client-side network is slow

Question 14

A team fine-tuned a Cohere Command R model using the T-Few technique on a dataset of JSONL prompt/completion pairs. After deployment, they observe that the model's responses are too repetitive. Which parameter adjustment in the OCI Generative AI Playground would BEST address this issue?

Accepted Answer

Increase frequency penalty. Increasing the frequency penalty discourages the model from repeating tokens that have already appeared, reducing repetitive outputs. Temperature and max tokens address randomness and length, not repetition specifically. Stop sequences are for terminating generation early.

Answer

Increase max tokens

Answer

Decrease temperature

Answer

Add a stop sequence

Question 15

Which THREE factors should be considered when choosing between fine-tuning a model and using a pre-trained model with prompt engineering? (Select three.)

Accepted Answer

Size of available dataset. The size of the available dataset is a critical factor: fine-tuning requires a sufficiently large, labeled dataset (typically thousands of examples) to adjust model weights effectively, while prompt engineering can work with zero or few examples. If the dataset is too small, fine-tuning risks overfitting and poor generalization, making prompt engineering the safer choice.

Answer

Required response time

Answer

Internet connectivity

Question 16

A data scientist in group DataScientists uses the OCI Generative AI SDK to start a fine-tuning job in compartment AIResources. They receive the error shown. What is the most likely cause?

Accepted Answer

The fine-tuning job requires additional IAM policies for accessing the training data in Object Storage.. The error message indicates a permissions issue related to accessing training data in Object Storage. When using the OCI Generative AI SDK to start a fine-tuning job, the data scientist's IAM policies must explicitly grant read access to the bucket and objects containing the training data. Without these policies, the API call fails even if the user is in the correct group and the compartment exists.

Answer

The compartment AIResources does not exist.

Answer

The fine-tuning API is not yet available in that region.

Answer

The data scientist is not in the DataScientists group.

Question 17

Which THREE models are available as part of the OCI Generative AI service?

Accepted Answer

Llama 3. Llama 3 is one of the open-source large language models (LLMs) available through the OCI Generative AI service, alongside Cohere models. OCI Generative AI provides managed access to Llama 3 for text generation tasks, allowing users to deploy and fine-tune it within Oracle Cloud Infrastructure.

Answer

GPT-4

Answer

Stable Diffusion

Question 18

A company is using OCI Generative AI to generate code snippets and notices that the model sometimes produces code with security vulnerabilities. They have a small dataset of secure code examples. Which approach would be most effective to reduce vulnerabilities?

Accepted Answer

Use prompt engineering with security constraints in the instruction.. Prompt engineering allows the company to inject security constraints directly into the instruction without requiring additional training data or infrastructure. By crafting a prompt that explicitly requests secure code (e.g., 'Generate code that follows OWASP Top 10 best practices and avoids SQL injection, XSS, and buffer overflows'), the model can leverage its existing knowledge to produce safer outputs. This approach is immediate, cost-effective, and does not depend on the size or quality of the small secure code dataset.

Answer

Use a different base model.

Answer

Fine-tune the model on the small secure code dataset.

Answer

Deploy a custom model hosted elsewhere.

Question 19

An enterprise is deploying a generative AI model that must comply with data residency regulations. Which two configurations should they implement? (Select TWO.)

Accepted Answer

Set up OCI IAM policies to prevent data egress from the region for the model's resources. OCI IAM policies can explicitly deny data egress from a specific region, ensuring that the generative AI model's resources (such as training data, model artifacts, and inference endpoints) remain within the region that satisfies data residency regulations. This is achieved by writing policy statements that restrict the movement of data across regional boundaries, which is a direct control for compliance.

Answer

Enable OCI Logging for all API calls

Answer

Use OCI Object Storage with cross-region replication for redundancy

Answer

Store encryption keys in an OCI Vault in a different region

Question 20

A prompt engineer is designing a ReAct pattern prompt to enable an LLM to use external tools. Which of the following is a key characteristic of the ReAct pattern?

Accepted Answer

The model produces a chain of thought followed by an action, then observes the result and continues reasoning. ReAct interleaves reasoning steps (e.g., 'I need to find the current weather') with actions (e.g., 'Action: call weather API') and observations, allowing the model to reason and act in a cycle.

Answer

The model uses a single tool call to answer the question

Answer

The model generates a final answer without any intermediate steps

Answer

The model is fine-tuned specifically for tool use