List your company
ResearchFormerly WizWand

Hallucination Detection on Summarization

32.1PRRAttention Score
4.95612.00319.0526.097May 26, 2025
Updated 4mo ago

Evaluation results

MethodLinks
Attention ScoreLLM Model=GPT-OSS 20B2025.0532.1
Attention ScoreLLM Model=Llama-3.1 8B-Instruct2025.0523
MSPLLM Model=GPT-OSS 20B2025.0522
RAUQLLM Model=GPT-OSS 20B2025.0519.6
Semantic EntropyLLM Model=GPT-OSS 20B2025.0519.4
MSPLLM Model=Llama-3.1 8B-Instruct2025.0518.4
MTELLM Model=GPT-OSS 20B2025.0518
CCPLLM Model=GPT-OSS 20B2025.0517.9
CCPLLM Model=Llama-3.1 8B-Instruct2025.0517.5
Semantic DensityLLM Model=GPT-OSS 20B2025.0516.8
PerplexityLLM Model=GPT-OSS 20B2025.0516.2
SARLLM Model=GPT-OSS 20B2025.0516.2
Semantic EntropyLLM Model=Llama-3.1 8B-Instruct2025.0515.2
RAUQLLM Model=Llama-3.1 8B-Instruct2025.0515
MTELLM Model=Llama-3.1 8B-Instruct2025.0514.4
Simple FocusLLM Model=GPT-OSS 20B2025.0514.2
PerplexityLLM Model=Llama-3.1 8B-Instruct2025.0513.9
SARLLM Model=Llama-3.1 8B-Instruct2025.0513.8
Semantic DensityLLM Model=Llama-3.1 8B-Instruct2025.0513.8
Lexical Similarity Rouge-LLLM Model=Llama-3.1 8B-Instruct2025.0512.8
LUQLLM Model=Llama-3.1 8B-Instruct2025.0512.1
Lexical Similarity Rouge-LLLM Model=GPT-OSS 20B2025.0512
Simple FocusLLM Model=Llama-3.1 8B-Instruct2025.0511.9
LUQLLM Model=GPT-OSS 20B2025.0510.7
EigenScoreLLM Model=Llama-3.1 8B-Instruct2025.0510.7
DegMat NLI Score entail.LLM Model=Llama-3.1 8B-Instruct2025.0510.1
EVL NLI Score entail.LLM Model=Llama-3.1 8B-Instruct2025.059.3
DegMat NLI Score entail.LLM Model=GPT-OSS 20B2025.059
EVL NLI Score entail.LLM Model=GPT-OSS 20B2025.058.6
EigenScoreLLM Model=GPT-OSS 20B2025.056.8
Ecc. NLI Score entail.LLM Model=GPT-OSS 20B2025.056.7
Ecc. NLI Score entail.LLM Model=Llama-3.1 8B-Instruct2025.056