List your company

Claim Verification on AVeriTeC Golden (dev)

83.4AccuracyDebateCV
60.5266.4672.478.34Jul 25, 2025

Evaluation results

MethodLinks
DebateCVLLMs=Llama-3.1-8B2025.0783.4
DebateCVLLMs=Qwen-2.5-7B2025.0782
MajorityLLMs=GPT-4o2025.0781.8
CooperationLLMs=GPT-4o2025.0781.8
CoTLLMs=GPT-4o2025.0781.6
CooperationLLMs=GPT-4o-Mini2025.0781.6
MajorityLLMs=GPT-4o-Mini2025.0781.4
InFactLLMs=GPT-4o2025.0781.2
CoTLLMs=GPT-4o-Mini2025.0780.8
HerOLLMs=Llama-3.1-8B2025.0780.4
HerOLLMs=Qwen-2.5-7B2025.0780
InFactLLMs=GPT-4o-Mini2025.0779.2
w/o Debate-SFTLLMs=GPT-4o2025.0778.8
MajorityLLMs=Llama-3.1-8B2025.0778
MajorityLLMs=Qwen-2.5-7B2025.0777.6
CooperationLLMs=Qwen-2.5-7B2025.0777.4
CoTLLMs=Qwen-2.5-7B2025.0777
CoTLLMs=Llama-3.1-8B2025.0776.8
w/o CcorrectLLMs=Llama-3.1-8B2025.0776.8
CooperationLLMs=Llama-3.1-8B2025.0776.6
w/o CerrorLLMs=Llama-3.1-8B2025.0776.4
w/o CerrorLLMs=Qwen-2.5-7B2025.0775.8
w/o CcorrectLLMs=Qwen-2.5-7B2025.0775.6
w/o Debate-SFTLLMs=GPT-4o-Mini2025.0774.6
InFactLLMs=Qwen-2.5-7B2025.0774.4
InFactLLMs=Llama-3.1-8B2025.0773.2
w/o Debate-SFTLLMs=Llama-3.1-8B2025.0766.6
w/o Debate-SFTLLMs=Qwen-2.5-7B2025.0761.4