List your company

Fact-Checking on AVERICITEC Retrieved Evidence (test)

77.7AccuracyRADARmulti
49.30856.67964.0571.421Apr 21, 2026

Evaluation results

MethodLinks
RADARmultiBackbone=GPT-4o mini2026.0477.763.345.656.587.6
RADARmultiBackbone=LLaMA3-8B-Instruct2026.0477.363.246.356.786.5
RADARmultiBackbone=Qwen2.5-7B-Instruct2026.0476.763.148.254.986.2
RADARsingleBackbone=Qwen2.5-7B-Instruct2026.0471.254.939.441.184.3
D2DBackbone=Qwen2.5-7B-Instruct2026.046948.522.241.282.1
D2DBackbone=LLaMA3-8B-Instruct2026.0464.145.121.734.179.5
FIREBackbone=Qwen2.5-7B-Instruct2026.0463.946.929.232.179.5
D2DBackbone=GPT-4o mini2026.046350.936.339.776.5
FIREBackbone=GPT-4o mini2026.0460.346.930.734.175.9
RADARsingleBackbone=GPT-4o mini2026.0458.45141.241.570.2
RADARsingleBackbone=LLaMA3-8B-Instruct2026.0457.35244.444.866.6
FIREBackbone=LLaMA3-8B-Instruct2026.0450.443.430.939.759.6