List your company

Language Modeling on Wikitext-2 (PPL)

5.47Perplexity (PPL)Dense
4.543610.796817.0523.3032May 28, 2026Jun 1, 2026Jun 6, 2026Jun 10, 2026Jun 15, 2026Jun 19, 2026Jun 24, 2026

Evaluation results

MethodLinks
DenseModel=LLaMA2-7B, Sparsity=Dense2026.065.47
RIAModel=LLaMA2-7B, Criterion=S-S, Sparsity=30%2026.065.64
DenseModel=LLaMA-7B, Sparsity=Dense2026.065.68
WandaModel=LLaMA2-7B, Criterion=S-S, Sparsity=30%2026.065.73
EPTSModel=LLaMA2-7B, Criterion=M-S, Sparsity=30%2026.065.74
SparseGPTModel=LLaMA2-7B, Criterion=S-S, Sparsity=30%2026.065.77
RIAModel=LLaMA-7B, Criterion=S-S, Sparsity=30%2026.065.86
EPTSModel=LLaMA-7B, Criterion=M-S, Sparsity=30%2026.065.95
SparseGPTModel=LLaMA-7B, Criterion=S-S, Sparsity=30%2026.065.96
RIAModel=LLaMA2-7B, Criterion=S-S, Sparsity=40%2026.065.96
WandaModel=LLaMA-7B, Criterion=S-S, Sparsity=30%2026.065.99
WandaModel=LLaMA2-7B, Criterion=S-S, Sparsity=40%2026.066.05
EPTSModel=LLaMA2-7B, Criterion=M-S, Sparsity=40%2026.066.07
SparseGPTModel=LLaMA2-7B, Criterion=S-S, Sparsity=40%2026.066.1
DenseModel=LLaMA3-8B, Sparsity=Dense2026.066.13
RIAModel=LLaMA-7B, Criterion=S-S, Sparsity=40%2026.066.21
SparseGPTModel=LLaMA-7B, Criterion=S-S, Sparsity=40%2026.066.32
EPTSModel=LLaMA-7B, Criterion=M-S, Sparsity=40%2026.066.33
WandaModel=LLaMA-7B, Criterion=S-S, Sparsity=40%2026.066.38
RIAModel=LLaMA3-8B, Criterion=S-S, Sparsity=30%2026.066.6
WandaModel=LLaMA3-8B, Criterion=S-S, Sparsity=30%2026.066.67
EPTSModel=LLaMA3-8B, Criterion=M-S, Sparsity=30%2026.066.67
SparseGPTModel=LLaMA3-8B, Criterion=S-S, Sparsity=30%2026.066.69
RIAModel=LLaMA2-7B, Criterion=S-S, Sparsity=50%2026.066.8
EPTSModel=LLaMA2-7B, Criterion=M-S, Sparsity=50%2026.066.86
WandaModel=LLaMA2-7B, Criterion=S-S, Sparsity=50%2026.066.92
SparseGPTModel=LLaMA2-7B, Criterion=S-S, Sparsity=50%2026.066.99
EPTSModel=LLaMA-7B, Criterion=M-S, Sparsity=50%2026.066.99
RIAModel=LLaMA-7B, Criterion=S-S, Sparsity=50%2026.067.13
SparseGPTModel=LLaMA-7B, Criterion=S-S, Sparsity=50%2026.067.19
WandaModel=LLaMA-7B, Criterion=S-S, Sparsity=50%2026.067.26
RIAModel=LLaMA3-8B, Criterion=S-S, Sparsity=40%2026.067.36
EPTSModel=LLaMA3-8B, Criterion=M-S, Sparsity=40%2026.067.43
WandaModel=LLaMA3-8B, Criterion=S-S, Sparsity=40%2026.067.44
SparseGPTModel=LLaMA3-8B, Criterion=S-S, Sparsity=40%2026.067.48
EPTSModel=LLaMA2-7B, Criterion=M-S, Sparsity=60%2026.068.52
EPTSModel=LLaMA-7B, Criterion=M-S, Sparsity=60%2026.068.64
RIAModel=LLaMA3-8B, Criterion=S-S, Sparsity=50%2026.069.33
SparseGPTModel=LLaMA3-8B, Criterion=S-S, Sparsity=50%2026.069.42
EPTSModel=LLaMA3-8B, Criterion=M-S, Sparsity=50%2026.069.48
WandaModel=LLaMA3-8B, Criterion=S-S, Sparsity=50%2026.069.82
SparseGPTModel=LLaMA2-7B, Criterion=S-S, Sparsity=60%2026.0610.13
SparseGPTModel=LLaMA-7B, Criterion=S-S, Sparsity=60%2026.0610.18
RIAModel=LLaMA2-7B, Criterion=S-S, Sparsity=60%2026.0610.4
RIAModel=LLaMA-7B, Criterion=S-S, Sparsity=60%2026.0610.63
WandaModel=LLaMA-7B, Criterion=S-S, Sparsity=60%2026.0610.73
WandaModel=LLaMA2-7B, Criterion=S-S, Sparsity=60%2026.0610.77
EPTSModel=LLaMA3-8B, Criterion=M-S, Sparsity=60%2026.0612.82
VanillaBackbone=Mistral-7B-Instruct2026.0513.9
DT-CD⋆Backbone=Mistral-7B-Instruct2026.0513.9
COFTBackbone=Mistral-7B-Instruct2026.0513.9
SDDBackbone=Mistral-7B-Instruct2026.0514.1
DExpertsBackbone=Mistral-7B-Instruct2026.0514.2
DenseModel=OPT-1.3B, Sparsity=Dense2026.0614.62
ICPModel=OPT-1.3B, Sparsity=30%2026.0614.67
WandaModel=OPT-1.3B, Criterion=S-S, Sparsity=30%2026.0615
ICPModel=OPT-1.3B, Sparsity=40%2026.0615.04
EPTSModel=OPT-1.3B, Criterion=M-S, Sparsity=30%2026.0615.1
SparseGPTModel=OPT-1.3B, Criterion=S-S, Sparsity=30%2026.0615.16
RIAModel=OPT-1.3B, Criterion=S-S, Sparsity=30%2026.0615.16
VanillaBackbone=LLaMA-2-13B2026.0515.3
DT-CD⋆Backbone=LLaMA-2-13B2026.0515.4
COFTBackbone=LLaMA-2-13B2026.0515.4
SparseGPTModel=LLaMA3-8B, Criterion=S-S, Sparsity=60%2026.0615.43
SDDBackbone=LLaMA-2-13B2026.0515.6
DExpertsBackbone=LLaMA-2-13B2026.0515.8
WandaModel=OPT-1.3B, Criterion=S-S, Sparsity=40%2026.0615.87
RIAModel=OPT-1.3B, Criterion=S-S, Sparsity=40%2026.0615.91
EPTSModel=OPT-1.3B, Criterion=M-S, Sparsity=40%2026.0616.03
EPTSModel=LLaMA2-7B, Criterion=M-S, Sparsity=70%2026.0616.47
ICPModel=OPT-1.3B, Sparsity=50%2026.0616.49
SparseGPTModel=OPT-1.3B, Criterion=S-S, Sparsity=40%2026.0616.5
EPTSModel=LLaMA-7B, Criterion=M-S, Sparsity=70%2026.0616.94
SparseGPTModel=OPT-1.3B, Criterion=S-S, Sparsity=50%2026.0617.45
EPTSModel=OPT-1.3B, Criterion=M-S, Sparsity=50%2026.0617.74
RIAModel=OPT-1.3B, Criterion=S-S, Sparsity=50%2026.0618.05
WandaModel=OPT-1.3B, Criterion=S-S, Sparsity=50%2026.0618.42
RIAModel=LLaMA3-8B, Criterion=S-S, Sparsity=60%2026.0619.64
EPTSModel=OPT-1.3B, Criterion=M-S, Sparsity=60%2026.0620.94
ICPModel=OPT-1.3B, Sparsity=60%2026.0621.17
DenseModel=OPT-350M, Sparsity=Dense2026.0622
SparseGPTModel=OPT-1.3B, Criterion=S-S, Sparsity=60%2026.0622.17
SparseGPTModel=OPT-350M, Criterion=S-S, Sparsity=30%2026.0623.19
WandaModel=OPT-350M, Criterion=S-S, Sparsity=30%2026.0623.49
EPTSModel=OPT-350M, Criterion=M-S, Sparsity=30%2026.0623.51
WandaModel=LLaMA3-8B, Criterion=S-S, Sparsity=60%2026.0623.52
RIAModel=OPT-350M, Criterion=S-S, Sparsity=30%2026.0623.54
EPTSModel=LLaMA3-8B, Criterion=M-S, Sparsity=70%2026.0625.02
SparseGPTModel=LLaMA-7B, Criterion=S-S, Sparsity=70%2026.0625.78
SparseGPTModel=OPT-350M, Criterion=S-S, Sparsity=40%2026.0625.8
RIAModel=OPT-1.3B, Criterion=S-S, Sparsity=60%2026.0626.28
WandaModel=OPT-1.3B, Criterion=S-S, Sparsity=60%2026.0626.49
WandaModel=OPT-350M, Criterion=S-S, Sparsity=40%2026.0626.54
RIAModel=OPT-350M, Criterion=S-S, Sparsity=40%2026.0626.55
EPTSModel=OPT-350M, Criterion=M-S, Sparsity=40%2026.0626.65
DenseModel=OPT-125M, Sparsity=Dense2026.0627.65
SparseGPTModel=LLaMA2-7B, Criterion=S-S, Sparsity=70%2026.0628.06
WandaModel=OPT-125M, Criterion=S-S, Sparsity=30%2026.0628.1
EPTSModel=OPT-125M, Criterion=M-S, Sparsity=30%2026.0628.37
RIAModel=OPT-125M, Criterion=S-S, Sparsity=30%2026.0628.63