List your company

Honesty Alignment on HonestyBench In-Domain

85.16NQ ScoreEliCal
55.655263.315170.97578.6349Oct 20, 2025

Evaluation results

MethodLinks
EliCalCategory=Upper Bound, Amount of annotated data=560k, Base Model=Qwen2.5-7B-Instruct2025.1085.1689.0986.0984.1988.8986.49
Cal-OnlyCategory=Upper Bound, Amount of annotated data=560k, Base Model=Qwen2.5-7B-Instruct2025.1084.8988.9685.6483.9788.0786.2
EliCalCategory=Training-based, Amount of annotated data=1k, Base Model=Qwen2.5-7B-Instruct2025.1082.3887.5184.4882.0584.3184.36
Consis-SemCategory=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1080.6890.280.1255.462.9373.62
Eli-OnlyCategory=Training-based, Base Model=Qwen2.5-7B-Instruct2025.1077.8686.2377.2754.3662.0571.19
Cal-OnlyCategory=Training-based, Amount of annotated data=1k, Base Model=Qwen2.5-7B-Instruct2025.1072.1968.7574.3476.1778.6173.41
Verbal-10Category=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1068.8262.3570.5373.2471.568.9
N-ProbCategory=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1066.1172.9661.9659.3361.6764.75
Consis-LexCategory=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1065.0274.9868.9867.8266.3569.8
Verbal-0Category=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1064.0270.2266.4965.0270.8167.22
DACACategory=Training-based, Base Model=Qwen2.5-7B-Instruct2025.1061.6972.5466.3965.0371.0667.7
ThermometerCategory=Training-based, Base Model=Qwen2.5-7B-Instruct2025.1058.1563.3858.0855.2467.5859.48
ProbCategory=Training-free, Base Model=Qwen2.5-7B-Instruct2025.1056.7970.2654.2941.7358.7155.48