Abstract: Assessing how Large Language Models (LLMs) acquire, retain, and retrieve factual information is crucial for understanding their reliability and knowledge depth. Existing benchmarks often ...