List your company

Hallucination Detection on CWQ (Resource Usage Metrics)

0LLM CallsKG-Guard(GraphTransformer)
-0.42.357.7May 29, 2026

Evaluation results

MethodLinks
KG-Guard(GraphTransformer)Encoder=all-roberta-large-v12026.0502.52358—
KG-Guard(GAT)Encoder=all-roberta-large-v12026.0501.84357—
Llama 4 Scout2026.051—109305
GPT 5 Mini2026.051———
SelfCheck(N=5)2026.055—109305
SelfCheck(N=10)2026.0510—109305