Benchmarking Truthfulness Metrics in Health-Oriented Large Language Models via Automated Fact Verification Pipelines. IJCHML 2026, 4 (2).