paper_type.json (381B)
1 { 2 "paper_type": "empirical", 3 "reason": "The paper runs systematic experiments on 10 LLMs using semantic-preserving mutations to empirically demonstrate that benchmark rankings are unstable and inflated, with performance drops up to 40-50%—the primary contribution is these experimental findings about benchmark vulnerability, not the dynamic benchmarking framework itself." 4 }