Final exam: Hallucination Detection and Verification: Catching Confident Wrong Answers

Quiz Summary

0 of 10 Questions completed

Questions:

You have already completed the quiz before. Hence you can not start it again.

Quiz is loading…

You must sign in or sign up to start the quiz.

You must first complete the following:

Results

Quiz complete. Results are being recorded.

Results

0 of 10 Questions answered correctly

Your time:

You have reached 0 of 0 point(s), (0)

Question 1 of 10

1. Question

An AI hallucination is defined as a system producing information that seems factual but is actually fabricated or incorrect.

a) A system that generates responses with high confidence but no factual basis
Correct Incorrect Correct answer

b) A system that mimics human speech patterns accurately
Correct Incorrect Correct answer

c) A system that requires no human oversight to function properly
Correct Incorrect Correct answer

d) A system that always provides correct and verified data
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 2 of 10

2. Question

Language models generate responses based on probabilistic patterns derived from training data rather than factual knowledge.

True, because models cannot verify facts or access external sources
Correct Incorrect Correct answer

False, because models are trained to always provide accurate information
Correct Incorrect Correct answer

True, but only when the training data is of high quality
Correct Incorrect Correct answer

False, because models have built-in fact-checking mechanisms
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 3 of 10

3. Question

What is one example of fabricated facts in a clinical setting according to the lesson?

Providing accurate patient symptoms from clinical records
Correct Incorrect Correct answer

Generating fictional clinical guidelines or diagnostic criteria
Correct Incorrect Correct answer

Citing quarterly earnings figures for a company that has not released those results yet
Correct Incorrect Correct answer

Assigning a theory to the correct academic researcher
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 4 of 10

4. Question

What distinguishes intrinsic hallucinations from extrinsic hallucinations in AI-generated responses?

Intrinsic hallucinations involve fabricated information not found in source documents, while extrinsic hallucinations use information from external sources.
Correct Incorrect Correct answer

Intrinsic hallucinations occur when AI references external frameworks, whereas extrinsic hallucinations create false facts within documents.
Correct Incorrect Correct answer

Intrinsic hallucinations are based on assumptions, while extrinsic hallucinations are derived from provided data.
Correct Incorrect Correct answer

Intrinsic hallucinations refer to missing data, while extrinsic hallucinations refer to incorrect data.
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 5 of 10

5. Question

What is the primary purpose of claim extraction according to the lesson?

To identify opinions within a response
Correct Incorrect Correct answer

To transform broad responses into testable statements
Correct Incorrect Correct answer

To eliminate all factual elements from an answer
Correct Incorrect Correct answer

To increase the length of written communication
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 6 of 10

6. Question

In natural language inference, entailment is defined as a relationship where the truth of one statement guarantees the truth of another.

Entailment occurs when two statements cannot both be true
Correct Incorrect Correct answer

Entailment is when a statement logically follows from another
Correct Incorrect Correct answer

Entailment exists when there is no clear logical connection between statements
Correct Incorrect Correct answer

Entailment happens when evidence directly opposes a claim
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 7 of 10

7. Question

Retrieval grounding involves tracing claims back to their original sources through systematic retrieval from trusted databases or documents.

True
Correct Incorrect Correct answer

False
Correct Incorrect Correct answer

Sometimes true
Correct Incorrect Correct answer

Not mentioned
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 8 of 10

8. Question

When verifying a quotation, what is the first step that should be taken to ensure accuracy?

Check the date of the original statement
Correct Incorrect Correct answer

Compare the quote against primary sources
Correct Incorrect Correct answer

Confirm the speaker's authority and position
Correct Incorrect Correct answer

Search through official records and press archives
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 9 of 10

9. Question

What is the primary purpose of checking self consistency in verification processes?

To ensure that external data matches internal records
Correct Incorrect Correct answer

To verify that information presented remains internally coherent
Correct Incorrect Correct answer

To confirm that all responses are completed within deadlines
Correct Incorrect Correct answer

To compare responses against industry benchmarks
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 10 of 10

10. Question

What is the primary goal of calibration in AI models according to the lesson?

To make models always provide the most confident responses possible
Correct Incorrect Correct answer

To align model confidence levels with actual accuracy
Correct Incorrect Correct answer

To eliminate all forms of uncertainty from model outputs
Correct Incorrect Correct answer

To reduce computational costs during inference
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points

Final exam: Hallucination Detection and Verification: Catching Confident Wrong Answers

Quiz Summary

Information

Results

Results

Categories

1. Question

2. Question

3. Question

4. Question

5. Question

6. Question

7. Question

8. Question

9. Question

10. Question