List your company

Token-level Hallucination Detection on AIME 2024 (AUROC/AUPRC)

87.39AUROCTOKENHD-8B
63.93870.026576.11582.2035May 12, 2026

Evaluation results

MethodLinks
TOKENHD-8B2026.0587.3973.59
o4-mini2026.0578.859.08
GPT-4.12026.0572.0644.26
QwQ-32B2026.0569.0449.19
R1-Qwen3-8B2026.0564.8445.62