HALLMARK: Human-centered Approach to Large Large Language Models for Accessing Reliable Knowledge