Veracity Inference on COMMONSENSEQA 1,000 examples
0.935Mean Hamming SimilarityVeracity Search (VS)
Evaluation Results
| Method | Links | |
|---|---|---|
| Veracity Search (VS)Base LM=Qwen-4B2025.05 | 0.935 | |
| Veracity Search (VS)Base LM=Qwen-8B2025.05 | 0.931 | |
| Veracity Search (VS)Base LM=Llama-8B2025.05 | 0.903 | |
| Veracity Search (VS)Base LM=Llama-3B2025.05 | 0.836 | |
| CoTBase LM=Qwen-4B2025.05 | 0.695 | |
| RecursiveBase LM=Qwen-4B2025.05 | 0.607 | |
| CoTBase LM=Qwen-8B2025.05 | 0.59 | |
| CoTBase LM=Llama-8B2025.05 | 0.535 | |
| Many2ManyBase LM=Qwen-8B2025.05 | 0.534 | |
| VotingBase LM=Qwen-8B2025.05 | 0.533 | |
| VotingBase LM=Qwen-4B2025.05 | 0.521 | |
| Many2ManyBase LM=Qwen-4B2025.05 | 0.517 | |
| RecursiveBase LM=Qwen-8B2025.05 | 0.509 | |
| CoTBase LM=Llama-3B2025.05 | 0.507 | |
| RecursiveBase LM=Llama-8B2025.05 | 0.506 | |
| RecursiveBase LM=Llama-3B2025.05 | 0.505 | |
| VotingBase LM=Llama-8B2025.05 | 0.505 | |
| Many2ManyBase LM=Llama-3B2025.05 | 0.504 | |
| VotingBase LM=Llama-3B2025.05 | 0.504 | |
| Many2ManyBase LM=Llama-8B2025.05 | 0.503 |