List your company

Hallucination Detection on HDMBench (test)

85HF1CALAMRFLOW
56.60863.97971.3578.721May 21, 2026

Evaluation results

MethodLinks
CALAMRFLOWAveraged over=5 random seeds2026.058576.296.151.274.653.4
CALAMRFLOWseeds=5, test-set size (N)=5202026.0585——51.274.6—
HDM-22026.0573.674.874.4—74.4—
HDM-2test-set size (N)=5202026.0573.6———74.4—
Qwen2026.0560.449.178.6—56.7—
Qwentest-set size (N)=5202026.0560.4———56.7—
GPT-4o2026.0558.768.451.4—69.6—
GPT-4otest-set size (N)=5202026.0558.7———69.6—
GPT-4o-mini2026.0557.768.449.9—69.3—
GPT-4o-minitest-set size (N)=5202026.0557.7———69.3—