AWS AI Practitioner
A company has built a content moderation system using a foundation model. As social media trends evolve rapidly, the company needs to keep the model updated with emerging harmful content patterns that reflect current trends. Which technique should the company use?
A
Prompt engineering
B
Continued pre-training
C
RAG with updated knowledge bases
D
Conduct reinforcement learning from human feedback (RLHF)
✓ Correcta
Explicación
Reinforcement Learning from Human Feedback (RLHF) is well-suited for content moderation systems that need to adapt to emerging trends. Human moderators can review model decisions on new content patterns, providing feedback signals that continuously improve the model's ability to identify evolving harmful content.