List your company

Hallucination Diagnosis on Hallucination Diagnosis Dataset Wikipedia-based

79.55ASQwen3-32B
58.978864.319469.6675.0006Dec 31, 2025

Evaluation results

MethodLinks
Qwen3-32BEvaluation Protocol=Pipeline Method2025.1279.5594.7476.9769.69
o4-miniEvaluation Protocol=Pipeline Method2025.1278.5182.4678.5979.71
GPT 4.1Evaluation Protocol=Pipeline Method2025.1276.7761.7560.5253.08
Qwen3-32BEvaluation Protocol=Single Prompt Method2025.1270.9897.5464.8559.15
HDM-4B-RLEvaluation Protocol=Single Prompt Method2025.1269.1692.2858.6548.49
GPT 4.1Evaluation Protocol=Single Prompt Method2025.1265.9777.5459.1248.1
o4-miniEvaluation Protocol=Single Prompt Method2025.1261.1889.1241.343.04
Original ResultDescription=Baseline for mitigation task2025.1259.77———