List your company

Fact Verification on FEVER (Acc, F1, FEVER Score)

87.45AccuracyInSemRAG
57.726865.443473.1680.8766May 31, 2026

Evaluation results

MethodLinks
InSemRAGLLM=DeepSeek2026.0587.4583.6593.55
SELF-RAGLLM=DeepSeek2026.0586.1382.0792.62
InSemRAGLLM=GPT2026.0585.1281.2591.05
IRCoTLLM=DeepSeek2026.0585.0976.5189.62
SELF-RAGLLM=GPT2026.0583.6279.6889.92
IRCoTLLM=GPT2026.0582.6174.2887.01
FiDLLM=DeepSeek2026.0582.4973.5388.9
InSemRAGLLM=Qwen2026.0581.1577.2586.85
FiDLLM=GPT2026.0580.0971.3986.31
ReActLLM=DeepSeek2026.0579.8571.4585.52
SELF-RAGLLM=Qwen2026.0579.4475.785.42
IRCoTLLM=Qwen2026.0578.4870.5782.66
ReActLLM=GPT2026.0577.5269.3783.03
Wikipedia GraphLLM=DeepSeek2026.0576.6269.3383.18
FiDLLM=Qwen2026.0576.0967.8281.99
Multi-Hop RAGLLM=DeepSeek2026.0575.8370.7384.46
Wikipedia GraphLLM=GPT2026.0574.3967.3180.76
ReActLLM=Qwen2026.0573.6465.978.88
Multi-Hop RAGLLM=GPT2026.0573.6268.6782
CoT-on-ConcatLLM=DeepSeek2026.0570.9868.3380.7
Wikipedia GraphLLM=Qwen2026.0570.6763.9476.72
Multi-Hop RAGLLM=Qwen2026.0569.9465.2477.9
CoT-on-ConcatLLM=GPT2026.0568.9166.3478.35
Naïve RAGLLM=DeepSeek2026.0568.8264.2679.49
Naïve RAGLLM=GPT2026.0566.8262.3977.17
CoT-on-ConcatLLM=Qwen2026.0565.4663.0274.43
w/o RAGLLM=DeepSeek2026.0563.8352.466.46
Naïve RAGLLM=Qwen2026.0563.4859.2773.31
w/o RAGLLM=GPT2026.0561.9750.8764.52
w/o RAGLLM=Qwen2026.0558.8748.3361.29