SpenseGPT introduces a hybrid sparse-dense weight format and one-shot pruning that delivers 1.2x end-to-end LLM decoding speedup on B200 GPUs with FP8 while preserving accuracy on Qwen3-32B and Seed-OSS-36B.
Beyond size: How gradients shape pruning decisions in large language models
4 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
4
Pith papers citing it
1
external citations · external index
fields
cs.LG 4representative citing papers
MaskPro learns categorical distributions over groups of M weights to generate exact (N:M) sparsity via N-way sampling without replacement and stabilizes training with a moving average tracker of loss residuals.
RAP is a reinforcement learning framework for runtime-adaptive pruning of LLMs that jointly optimizes model weights and KV-cache usage under varying memory budgets.
citing papers explorer
-
SpenseGPT: Practical One-shot Pruning Enabling Sparse and Dense GEMMs for LLM Inference
SpenseGPT introduces a hybrid sparse-dense weight format and one-shot pruning that delivers 1.2x end-to-end LLM decoding speedup on B200 GPUs with FP8 while preserving accuracy on Qwen3-32B and Seed-OSS-36B.
-
MaskPro: Linear-Space Probabilistic Learning for Strict (N:M)-Sparsity on LLMs
MaskPro learns categorical distributions over groups of M weights to generate exact (N:M) sparsity via N-way sampling without replacement and stabilizes training with a moving average tracker of loss residuals.
-
RAP: Runtime Adaptive Pruning for LLM Inference
RAP is a reinforcement learning framework for runtime-adaptive pruning of LLMs that jointly optimizes model weights and KV-cache usage under varying memory budgets.
- RT-Lynx: Putting GEMM Sparsity in the Right Place for Diffusion Models