List your company

Hallucination Detection on Company

0.93AUC-ROCLatent Debate Detector
0.47240.59120.710.8288May 30, 2025Jun 29, 2025Jul 30, 2025Aug 30, 2025Sep 30, 2025Oct 31, 2025Dec 1, 2025

Evaluation results

MethodLinks
Latent Debate Detector2025.120.93
SAPLMA2025.120.88
CDESLLM=Vicuna-13B2025.050.798
SDESLLM=LLama-2-13B2025.050.784
CDESLLM=Alpaca-13B2025.050.753
SDESLLM=LLama-2-7B2025.050.738
CDESLLM=Gemma-2-9B2025.050.736
ODESLLM=Vicuna-13B2025.050.729
CDESLLM=LLama-2-13B2025.050.728
SDESLLM=Gemma-2-9B2025.050.728
SDESLLM=Vicuna-13B2025.050.723
VarInit2025.120.72
SelfCheckGPT2025.120.72
ODESLLM=Gemma-2-9B2025.050.714
SDESLLM=Mistral-7B-0.32025.050.713
AvgInit2025.120.71
SDESLLM=Alpaca-13B2025.050.705
MINDLLM=Vicuna-13B2025.050.698
SAPLMALLM=Vicuna-13B2025.050.682
ODESLLM=Alpaca-13B2025.050.678
Probe@ExactLLM=Vicuna-13B2025.050.672
CDESLLM=Mistral-7B-0.32025.050.669
VarFin2025.120.66
MINDLLM=Gemma-2-9B2025.050.659
CDESLLM=LLama-2-7B2025.050.659
ODESLLM=LLama-2-13B2025.050.653
SAPLMALLM=Gemma-2-9B2025.050.648
Probe@ExactLLM=Mistral-7B-0.32025.050.644
AvgFin2025.120.64
Probe@ExactLLM=Gemma-2-9B2025.050.639
ODESLLM=Mistral-7B-0.32025.050.635
SAPLMALLM=Mistral-7B-0.32025.050.632
MINDLLM=Alpaca-13B2025.050.624
Probe@ExactLLM=Alpaca-13B2025.050.612
Probe@ExactLLM=LLama-2-13B2025.050.607
MINDLLM=LLama-2-13B2025.050.603
AvgProbLLM=Vicuna-13B2025.050.602
MINDLLM=Mistral-7B-0.32025.050.601
ODESLLM=LLama-2-7B2025.050.597
AvgProbLLM=Mistral-7B-0.32025.050.593
SAPLMALLM=Alpaca-13B2025.050.593
AvgProbLLM=LLama-2-13B2025.050.592
AvgProbLLM=LLama-2-7B2025.050.59
SAPLMALLM=LLama-2-13B2025.050.582
AvgProbLLM=Gemma-2-9B2025.050.58
AvgEntLLM=LLama-2-13B2025.050.564
MINDLLM=LLama-2-7B2025.050.564
AvgEntLLM=Mistral-7B-0.32025.050.562
Probe@ExactLLM=LLama-2-7B2025.050.559
EUBHDLLM=Gemma-2-9B2025.050.556
EUBHDLLM=Vicuna-13B2025.050.553
AvgEntLLM=Alpaca-13B2025.050.542
AvgEntLLM=Gemma-2-9B2025.050.541
EUBHDLLM=Alpaca-13B2025.050.541
AvgEntLLM=LLama-2-7B2025.050.54
SAPLMALLM=LLama-2-7B2025.050.54
EUBHDLLM=Mistral-7B-0.32025.050.538
AvgEntLLM=Vicuna-13B2025.050.533
EUBHDLLM=LLama-2-13B2025.050.532
AvgProbLLM=Alpaca-13B2025.050.53
EUBHDLLM=LLama-2-7B2025.050.525
P(True)LLM=Vicuna-13B2025.050.524
P(True)LLM=Mistral-7B-0.32025.050.515
P(True)LLM=LLama-2-7B2025.050.514
P(True)LLM=Gemma-2-9B2025.050.51
P(True)LLM=Alpaca-13B2025.050.506
P(True)LLM=LLama-2-13B2025.050.491
NumAtk2025.120.49