List your company

Veracity Inference on PRONTOQA (1,000 examples)

96.4Mean Hamming SimilarityVeracity Search (VS)
36.0851.7467.483.06May 17, 2025

Evaluation results

MethodLinks
Veracity Search (VS)Base LM=Llama-8B2025.0596.4
Veracity Search (VS)Base LM=Llama-3B2025.0594.8
Veracity Search (VS)Base LM=Qwen-8B2025.0594.5
Veracity Search (VS)Base LM=Qwen-4B2025.0591
VotingBase LM=Qwen-8B2025.0569.2
RecursiveBase LM=Qwen-4B2025.0569.1
Many2ManyBase LM=Qwen-8B2025.0568.3
RecursiveBase LM=Qwen-8B2025.0566.7
VotingBase LM=Qwen-4B2025.0560.3
CoTBase LM=Qwen-4B2025.0559.1
Many2ManyBase LM=Qwen-4B2025.0559
RecursiveBase LM=Llama-3B2025.0553.8
VotingBase LM=Llama-8B2025.0553.6
Many2ManyBase LM=Llama-8B2025.0553
CoTBase LM=Llama-8B2025.0551.5
VotingBase LM=Llama-3B2025.0551.4
Many2ManyBase LM=Llama-3B2025.0550.6
RecursiveBase LM=Llama-8B2025.0547.1
CoTBase LM=Llama-3B2025.0545.9
CoTBase LM=Qwen-8B2025.0538.4