List your company

Contamination Evaluation on ArxivRollBench

0.48Absolute RS_ILlama3.1-70B
0.44280.69390.9451.1961Jul 25, 2025

Evaluation results

MethodLinks
Llama3.1-70B2025.070.480.02
Phi-22025.070.620.36
Llama3.1-8B2025.070.670.25
Qwen2-7B2025.070.690.42
Qwen2.5-7B2025.070.70.37
Llama3-8B2025.070.740.74
Llama-3.1-Nemotron-70B2025.070.770.14
Yi-1.5-34B2025.070.810.55
Phi-1.52025.070.820.57
Phi-3.5-mini2025.071.070.07
Llama2-7B2025.071.140.11
Phi-12025.071.211.31
Phi-3-mini2025.071.270.05
Qwen2.5-72B2025.071.410.32