List your company

Veracity Inference on GSM8K 1,000 examples

75.1Mean Hamming SimilarityVeracity Search (VS)
48.5855.46562.3569.235May 17, 2025

Evaluation results

MethodLinks
Veracity Search (VS)Base LM=Qwen-8B2025.0575.1
Veracity Search (VS)Base LM=Qwen-4B2025.0571.1
CoTBase LM=Qwen-8B2025.0569.5
VotingBase LM=Qwen-8B2025.0565.4
Many2ManyBase LM=Qwen-8B2025.0565
Veracity Search (VS)Base LM=Llama-8B2025.0564.6
VotingBase LM=Qwen-4B2025.0562.3
Many2ManyBase LM=Qwen-4B2025.0562
RecursiveBase LM=Qwen-8B2025.0561.7
CoTBase LM=Qwen-4B2025.0561.4
Veracity Search (VS)Base LM=Llama-3B2025.0561.4
RecursiveBase LM=Llama-3B2025.0556.8
RecursiveBase LM=Llama-8B2025.0556.8
Many2ManyBase LM=Llama-8B2025.0556.7
Many2ManyBase LM=Llama-3B2025.0556.6
VotingBase LM=Llama-3B2025.0556.6
VotingBase LM=Llama-8B2025.0555.3
RecursiveBase LM=Qwen-4B2025.0554
CoTBase LM=Llama-3B2025.0549.6
CoTBase LM=Llama-8B2025.0549.6