Share your thoughts, 1 month free Claude Pro on usSee more

Claim Verification on AVeriTeC Golden (dev)

83.4Accuracy

DebateCV

Updated 1mo ago

Evaluation Results

Method	Links
DebateCV 2025.07		83.4
DebateCV 2025.07		82
Majority 2025.07		81.8
Cooperation 2025.07		81.8
CoT 2025.07		81.6
Cooperation 2025.07		81.6
Majority 2025.07		81.4
InFact 2025.07		81.2
CoT 2025.07		80.8
HerO 2025.07		80.4
HerO 2025.07		80
InFact 2025.07		79.2
w/o Debate-SFT 2025.07		78.8
Majority 2025.07		78
Majority 2025.07		77.6
Cooperation 2025.07		77.4
CoT 2025.07		77
CoT 2025.07		76.8
w/o Ccorrect 2025.07		76.8
Cooperation 2025.07		76.6
w/o Cerror 2025.07		76.4
w/o Cerror 2025.07		75.8
w/o Ccorrect 2025.07		75.6
w/o Debate-SFT 2025.07		74.6
InFact 2025.07		74.4
InFact 2025.07		73.2
w/o Debate-SFT 2025.07		66.6
w/o Debate-SFT 2025.07		61.4