List your company

Hallucination Detection on HaluEval (test)

98.55AUC-ROCMultiHaluDet
62.919672.169881.4290.6702May 30, 2025Jul 30, 2025Sep 29, 2025Nov 29, 2025Jan 29, 2026Mar 31, 2026May 31, 2026

Evaluation results

MethodLinks
MultiHaluDetLLM Backbone=LLaMA2-7B2026.0598.55———————
MultiHaluDetLLM Backbone=Mistral-7B2026.0598.43———————
CDESLLM Backbone=Llama-2-13B2025.0597.1———————
SDESLLM Backbone=Alpaca-13B2025.0597.1———————
CDESLLM Backbone=Vicuna-13B2025.0596.9———————
CDESLLM Backbone=Gemma-2-9B2025.0596———————
ODESLLM Backbone=Vicuna-13B2025.0595.4———————
CDESLLM Backbone=Mistral-7B-v0.32025.0595.4———————
SDESLLM Backbone=Llama-2-13B2025.0595.4———————
Neural CDEsLLM Backbone=Mistral-7B2026.0595.4———————
CDESLLM Backbone=Alpaca-13B2025.0595.3———————
CALAMRFLOWAveraged over=5 random seeds2026.0594.9——96.493.394.894.994.9
MINDLLM Backbone=Mistral-7B-v0.32025.0594.5———————
MINDLLM Backbone=Mistral-7B2026.0594.5———————
Probe@ExactLLM Backbone=Vicuna-13B2025.0594.1———————
ODESLLM Backbone=Llama-2-13B2025.0593.9———————
MINDLLM Backbone=Llama-2-13B2025.0593.8———————
MINDLLM Backbone=Alpaca-13B2025.0593.7———————
SDESLLM Backbone=Mistral-7B-v0.32025.0593.7———————
Neural SDEsLLM Backbone=Mistral-7B2026.0593.7———————
Probe@ExactLLM Backbone=Alpaca-13B2025.0593.5———————
Probe@ExactLLM Backbone=Mistral-7B-v0.32025.0593.4———————
Probe@ExactLLM Backbone=Mistral-7B2026.0593.4———————
SAPLMALLM Backbone=Vicuna-13B2025.0593.1———————
SDESLLM Backbone=Vicuna-13B2025.0593.1———————
MINDLLM Backbone=Vicuna-13B2025.0592.9———————
SDESLLM Backbone=Llama-2-7B2025.0592.8———————
Neural SDEsLLM Backbone=LLaMA2-7B2026.0592.8———————
TriLensLLM=Gemma-2-9B2026.0592.77———————
SDESLLM Backbone=Gemma-2-9B2025.0592.6———————
Probe@ExactLLM Backbone=Llama-2-13B2025.0592.4———————
Probe@ExactLLM Backbone=Gemma-2-9B2025.0592.1———————
ODESLLM Backbone=Alpaca-13B2025.0592.1———————
CDESLLM Backbone=Llama-2-7B2025.0591.4———————
Neural CDEsLLM Backbone=LLaMA2-7B2026.0591.4———————
TriLensLLM=Llama-3-8B2026.0591.36———————
ODESLLM Backbone=Mistral-7B-v0.32025.0591.2———————
Neural ODEsLLM Backbone=Mistral-7B2026.0591.2———————
MINDLLM Backbone=Gemma-2-9B2025.0591———————
TriLensLLM=Qwen2.5-7B2026.0590.53———————
SAPLMALLM Backbone=Gemma-2-9B2025.0590.5———————
ODESLLM Backbone=Gemma-2-9B2025.0590.5———————
SAPLMALLM Backbone=Llama-2-13B2025.0590.1———————
SAPLMALLM Backbone=Alpaca-13B2025.0589.5———————
ODESLLM Backbone=Llama-2-7B2025.0589.5———————
Neural ODEsLLM Backbone=LLaMA2-7B2026.0589.5———————
SAPLMALLM Backbone=Mistral-7B-v0.32025.0589.4———————
SAPLMALLM Backbone=Mistral-7B2026.0589.4———————
Probe@ExactLLM Backbone=Llama-2-7B2025.0588.3———————
Probe@ExactLLM Backbone=LLaMA2-7B2026.0588.3———————
SAPLMALLM Backbone=Llama-2-7B2025.0587———————
SAPLMALLM Backbone=LLaMA2-7B2026.0587———————
MINDLLM Backbone=Llama-2-7B2025.0586.1———————
MINDLLM Backbone=LLaMA2-7B2026.0586.1———————
ICR ProbeLLM=Gemma-2-9B2026.0584.36———————
SAPLMALLM=Gemma-2-9B2026.0581.23———————
HALLUGUARDBackbone=Llama2-70B2026.0180.7979.5468.51—————
ICR ProbeLLM=Qwen2.5-7B2026.0580.76———————
HALLUGUARDBackbone=Llama2-13B2026.0178.1574.1561.14—————
EUBHDLLM Backbone=Llama-2-13B2025.0578.1———————
SAPLMALLM=Qwen2.5-7B2026.0577.25———————
INSIDELLM Backbone=Llama-2-13B2025.0576.9———————
ICR ProbeLLM=Llama-3-8B2026.0576.71———————
EUBHDLLM Backbone=Vicuna-13B2025.0576———————
INSIDELLM Backbone=Mistral-7B-v0.32025.0576———————
INSIDELLM Backbone=Mistral-7B2026.0576———————
INSIDELLM Backbone=Gemma-2-9B2025.0575.8———————
HALLUGUARDBackbone=Llama2-7B2026.0175.7272.8963.15—————
INSIDELLM Backbone=Vicuna-13B2025.0575.2———————
InsideBackbone=Llama2-70B2026.0174.6268.3364.4—————
INSIDELLM Backbone=Llama-2-7B2025.0574.5———————
INSIDELLM Backbone=LLaMA2-7B2026.0574.5———————
LN-EntropyLLM=Gemma-2-9B2026.0573.57———————
HALLUGUARDBackbone=Llama-3-8B2026.0173.4371.1954.8—————
INSIDELLM Backbone=Alpaca-13B2025.0573.3———————
SAPLMALLM=Llama-3-8B2026.0573.15———————
SEPLLM=Llama-3-8B2026.0573.09———————
LN-EntropyLLM=Qwen2.5-7B2026.0572.86———————
EUBHDLLM Backbone=Gemma-2-9B2025.0572.6———————
InsideBackbone=Llama2-13B2026.0172.0171.9760.64—————
EUBHDLLM Backbone=Llama-2-7B2025.0571.9———————
EUBHDLLM Backbone=LLaMA2-7B2026.0571.9———————
InsideBackbone=Llama2-7B2026.0171.3367.6353.15—————
EUBHDLLM Backbone=Alpaca-13B2025.0571.3———————
EUBHDLLM Backbone=Mistral-7B-v0.32025.0570.5———————
EUBHDLLM Backbone=Mistral-7B2026.0570.5———————
InsideBackbone=Llama-3-8B2026.0167.9564.9352.21—————
SEPLLM=Qwen2.5-7B2026.0566.34———————
P(true)Backbone=Llama2-13B2026.0165.7763.0145.47—————
LN-EntropyLLM=Llama-3-8B2026.0565.74———————
SelfCheckGPTBackbone=Llama-3-8B2026.0165.4463.1348.23—————
SelfCheckGPTBackbone=Llama2-13B2026.0165.2463.5254.33—————
LN-EntropyBackbone=Llama2-13B2026.0165.1863.5348.09—————
FActScoreBackbone=Llama2-13B2026.0165.1563.7154.61—————
LN-EntropyBackbone=Llama-3-8B2026.0164.8960.7246.39—————
MINDBackbone=Llama-3-8B2026.0164.5460.8945.13—————
NLILLM Backbone=Gemma-2-9B2025.0564.4———————
SEPLLM=Gemma-2-9B2026.0564.39———————
Lexical Sim.Backbone=Llama2-70B2026.0164.3760.9250.86—————
SelfCheckGPTBackbone=Llama2-7B2026.0164.2961.8345.49—————