List your company

Claim Verification on AVeriTeC Retrieved (H) (dev)

72.8AccuracyDebateCV
53.0458.1763.368.43Jul 25, 2025

Evaluation results

MethodLinks
DebateCVLLMs=Llama-3.1-8B2025.0772.854.4
DebateCVLLMs=Qwen-2.5-7B2025.0772.654.6
w/o CcorrectLLMs=Llama-3.1-8B2025.0770.853.8
HerOLLMs=Llama-3.1-8B2025.0770.252.6
w/o CcorrectLLMs=Qwen-2.5-7B2025.0769.453
InFactLLMs=GPT-4o2025.0768.652.6
HerOLLMs=Qwen-2.5-7B2025.0768.250.6
CooperationLLMs=GPT-4o2025.0767.651.2
w/o Debate-SFTLLMs=GPT-4o2025.0765.448.6
InFactLLMs=GPT-4o-Mini2025.0765.247.8
CooperationLLMs=GPT-4o-Mini2025.0765.249.4
MajorityLLMs=GPT-4o2025.0764.449.2
CoTLLMs=GPT-4o2025.0763.847.8
MajorityLLMs=GPT-4o-Mini2025.0763.847.8
MajorityLLMs=Llama-3.1-8B2025.0763.847.6
InFactLLMs=Qwen-2.5-7B2025.0763.647.4
MajorityLLMs=Qwen-2.5-7B2025.0763.648.2
InFactLLMs=Llama-3.1-8B2025.0763.447
w/o CerrorLLMs=Llama-3.1-8B2025.076349.2
w/o CerrorLLMs=Qwen-2.5-7B2025.0762.949.1
CoTLLMs=GPT-4o-Mini2025.0762.846.8
CoTLLMs=Qwen-2.5-7B2025.0762.646.8
CoTLLMs=Llama-3.1-8B2025.0762.446.6
CooperationLLMs=Llama-3.1-8B2025.0761.246.8
CooperationLLMs=Qwen-2.5-7B2025.0760.846.4
w/o Debate-SFTLLMs=GPT-4o-Mini2025.0754.840.4
w/o Debate-SFTLLMs=Qwen-2.5-7B2025.0754.241
w/o Debate-SFTLLMs=Llama-3.1-8B2025.0753.841.2