Natural Language Inference Distribution Estimation on ChaosNLI (test)
0.06KL DivergenceChaosNLI HJD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ChaosNLI HJDLLM=Baseline, Fine-tuning Evaluator Model=RoBERTa2024.12 | 0.06 | 0.922 | 65.3 | |
| ChaosNLI HJDLLM=Baseline, Fine-tuning Evaluator Model=BERT2024.12 | 0.077 | 0.974 | 60.9 | |
| Llama3 Label-FreeLLM=Llama3, Explanation Type=Label-Free, Fine-tuning Evaluator Model=RoBERTa2024.12 | 0.105 | 1.058 | 57.1 | |
| Llama3 Label-FreeLLM=Llama3, Explanation Type=Label-Free, Fine-tuning Evaluator Model=BERT2024.12 | 0.107 | 1.063 | 53.3 |