ResearchBenchmarksLLM Pruning Evaluation on Full Benchmark SuiteFollow75.15S_tot RetentionSparseGPT64.500467.265270.0372.7948Jun 2, 2026Evaluation ResultsMethodMethodLinksS_tot RetentionSparseGPTCalibration=mix_balanc...Calibration=mix_balanced, Model Scale=LLaMA-3.1-70B, Sparsity=60% unstructured2026.0675.15WandaCalibration=IGSP (ours...Calibration=IGSP (ours), Model Scale=LLaMA-3.1-70B, Sparsity=60% unstructured2026.0664.91