Correctness Prediction on Cities
88AUROCDirection
Evaluation Results
| Method | Links | |
|---|---|---|
| DirectionModel=Llama 3.3 70B Instruct, Train dataset=TriviaQA2025.09 | 88 | |
| DirectionModel=Mistral 7B Instruct v0.3, Train dataset=TriviaQA2025.09 | 88 | |
| DirectionModel=DeepSeek R1 Distill Qwen 32B, Train dataset=TriviaQA2025.09 | 87.9 | |
| DirectionModel=Qwen 2.5 7B Instruct, Train dataset=TriviaQA2025.09 | 84.2 | |
| DirectionModel=Ministral 8B Instruct 2410, Train dataset=TriviaQA2025.09 | 84 | |
| Verb. conf.Model=Qwen 2.5 7B Instruct2025.09 | 75.8 | |
| DirectionModel=Llama 3.1 8B, Train dataset=TriviaQA2025.09 | 73.2 | |
| AssessorModel=Mistral 7B Instruct v0.3, Assessor type=logistic regression2025.09 | 71 | |
| AssessorModel=Qwen 2.5 7B Instruct, Assessor type=logistic regression2025.09 | 70.8 | |
| Verb. conf.Model=Mistral 7B Instruct v0.32025.09 | 70.5 | |
| Verb. conf.Model=Llama 3.3 70B Instruct2025.09 | 69.4 | |
| AssessorModel=Ministral 8B Instruct 2410, Assessor type=logistic regression2025.09 | 68.2 | |
| AssessorModel=Llama 3.3 70B Instruct, Assessor type=logistic regression2025.09 | 67.2 | |
| AssessorModel=Llama 3.1 8B, Assessor type=logistic regression2025.09 | 66.3 | |
| AssessorModel=DeepSeek R1 Distill Qwen 32B, Assessor type=logistic regression2025.09 | 66.3 | |
| Verb. conf.Model=DeepSeek R1 Distill Qwen 32B2025.09 | 57.7 | |
| Verb. conf.Model=Ministral 8B Instruct 24102025.09 | 55.4 | |
| Verb. conf.Model=Llama 3.1 8B2025.09 | 50 |