List your company

Truthfulness Evaluation on TruthfulQA (BLEU-Acc)

54.7BLEU-AccSG
36.29241.07145.8550.629Oct 10, 2025

Evaluation results

MethodLinks
SGBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers, Decision Layer=smooth (sigmoid)2025.1054.7
ER-DBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1052.1
DBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1052
ER-GBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1052
SCBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1052
SGBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers, Decision Layer=smooth (sigmoid)2025.1051.4
MIBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1051.4
GBackbone=GPT-OSS-20B, Candidate Generation=k=10 short answers2025.1051.2
SCBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1051.1
SGBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers, Decision Layer=smooth (sigmoid)2025.1050.9
SGBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers, Decision Layer=smooth (sigmoid)2025.1050.4
SGBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers, Decision Layer=smooth (sigmoid)2025.1050.3
GBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1050.1
MIBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1049.8
ER-GBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1048.8
MIBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1048.6
GBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1048.5
MIBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1047.8
GBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1047.8
ER-DBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1046.7
SCBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1046.6
GBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1046.2
SCBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1046
MIBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1046
DBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1045.6
SCBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1045.2
ER-GBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1045
ER-DBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1044.7
DBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1044.5
ER-GBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1044.4
ER-GBackbone=Llama-3.1-8B, Candidate Generation=k=10 short answers2025.1044.4
ER-DBackbone=Llama-3.2-1B, Candidate Generation=k=10 short answers2025.1044.4
DBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1044
ER-DBackbone=LLaMA-2-7B, Candidate Generation=k=10 short answers2025.1043.5
DBackbone=LLaMA-2-13B, Candidate Generation=k=10 short answers2025.1037