Bias Mitigation & Fairness — NCA-GENL Practice Question
A representative NVIDIA Generative AI LLMs Associate (NCA-GENL) exam question on Bias Mitigation & Fairness. Work through it below, then read why each option is right or wrong.
Short answer
The correct answer is D. Audit the training data for representation gaps across dialects and augment underrepresented linguistic patterns.
Auditing the training data for representation gaps and augmenting underrepresented dialects directly addresses the root cause of the bias, which is skewed training data. Removing demographic information (A) does not fix the underlying linguistic bias since dialect patterns persist even without explicit demographic labels. Adding a disclaimer (B) acknowledges the issue but does not mitigate it. Fine-tuning only on standardized terminology (C) would worsen the problem by making the model even less capable of understanding non-standard inputs.
The Question
A healthcare startup is deploying an LLM-based triage assistant that recommends urgency levels for patient complaints. During testing, the team discovers the model consistently assigns lower urgency scores to complaints written in non-standard English dialects. What is the most effective first step to address this bias?
Why D is correct
Auditing the training data for representation gaps and augmenting underrepresented dialects directly addresses the root cause of the bias, which is skewed training data. Removing demographic information (A) does not fix the underlying linguistic bias since dialect patterns persist even without explicit demographic labels. Adding a disclaimer (B) acknowledges the issue but does not mitigate it. Fine-tuning only on standardized terminology (C) would worsen the problem by making the model even less capable of understanding non-standard inputs.
Why the other options are wrong
Option A does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.
Option B does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.
Option C does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.
Key idea: Bias Mitigation & Fairness
Auditing the training data for representation gaps and augmenting underrepresented dialects directly addresses the root cause of the bias, which is skewed training data. Removing demographic information (A) does not fix the underlying linguistic bias since dialect patterns persist even without explicit demographic labels. Adding a disclaimer (B) acknowledges the issue but does not mitigate it. Fine-tuning only on standardized terminology (C) would worsen the problem by making the model even less capable of understanding non-standard inputs. On the NCA-GENL exam, questions in the "Trustworthy AI" domain test whether you can map a scenario's constraints to the right choice. Read the requirement carefully, eliminate options that violate any single constraint, and pick the one that satisfies all of them with the least operational overhead.
Ready to see how you'd score?
Take the free practice quiz and find out which NVIDIA Generative AI LLMs Associate domains you need to focus on. No signup required.
Practice 5 similar questions
Same cert, same or adjacent domain. Use these after reviewing the explanation.