Hallucination Detection

TruthfulQA is designed to catch a specific model failure. Describe what it tests and why larger models sometimes do worse.

Describe a concrete, production runnable mechanism to detect hallucinations in a RAG answer (claims unsupported by the retrieved chunks), as the answer is generated or just after.

Describe two automated hallucination detection techniques and their tradeoffs

What is the main limitation of self-check consistency as a hallucination detection technique?

Explain FActScore's approach to long-form factual evaluation and its design decisions

What does FActScore measure and how does it decompose long-form factual evaluation?