Hallucination Detection on HaluEval QA
99.5AccuracyAbnorDetector-Full
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AbnorDetector-FullBase LLM=Llama-3.1-8B-Instruct, alpha (α)=0.25, beta (β)=0.252025.04 | 99.5 | 0 | 99.5 | |
| AbnorDetector-FullBase LLM=Llama-3.1-70B-Instruct, alpha (α)=0.25, beta (β)=0.252025.04 | 98.5 | 0 | 98.5 | |
| AbnorDetector-FullBase LLM=Llama-3.1-70B-Instruct, alpha (α)=0.125, beta (β)=0.1252025.04 | 98.5 | 0 | 98.5 | |
| AbnorDetector-FullBase LLM=Llama-2-7b-chat, alpha (α)=0.25, beta (β)=0.252025.04 | 98 | 0 | 98 | |
| AbnorDetector-FullBase LLM=Gemma-7b-it, alpha (α)=0.25, beta (β)=0.252025.04 | 98 | 0 | 98 | |
| AbnorDetector-FullBase LLM=Llama-2-7b-chat, alpha (α)=0.125, beta (β)=0.1252025.04 | 97.83 | 0.0556 | 97.83 | |
| AbnorDetector-FullBase LLM=Gemma-7b-it, alpha (α)=0.125, beta (β)=0.1252025.04 | 97.5 | 0 | 97.5 | |
| AbnorDetector-LiteBase LLM=Llama-3.1-70B-Instruct, alpha (α)=0.5, beta (β)=0.52025.04 | 96.33 | 0.0556 | 96.33 | |
| AbnorDetector-FullBase LLM=Llama-3.1-8B-Instruct, alpha (α)=0.125, beta (β)=0.1252025.04 | 95.5 | 0 | 95.5 | |
| AbnorDetector-LiteBase LLM=Gemma-7b-it, alpha (α)=0.25, beta (β)=0.252025.04 | 94.33 | 2.0556 | 94.33 | |
| AbnorDetector-LiteBase LLM=Llama-3.1-8B-Instruct, alpha (α)=0.5, beta (β)=0.52025.04 | 93.5 | 0.5 | 93.49 | |
| AbnorDetector-LiteBase LLM=Gemma-7b-it, alpha (α)=0.5, beta (β)=0.52025.04 | 93.5 | 1.1667 | 93.49 | |
| AbnorDetector-LiteBase LLM=Llama-2-7b-chat, alpha (α)=0.5, beta (β)=0.52025.04 | 92.83 | 0.3889 | 92.82 | |
| AbnorDetector-LiteBase LLM=Llama-3.1-70B-Instruct, alpha (α)=0.25, beta (β)=0.252025.04 | 92.17 | 3.7222 | 92.16 | |
| AbnorDetector-LiteBase LLM=Llama-3.1-8B-Instruct, alpha (α)=0.25, beta (β)=0.252025.04 | 86.67 | 0.0556 | 86.57 | |
| Lynx2025.04 | 86.1 | — | 85.41 | |
| AbnorDetector-LiteBase LLM=Llama-2-7b-chat, alpha (α)=0.25, beta (β)=0.252025.04 | 82.83 | 5.7222 | 82.51 |