conftrace_

pruning

33 papers

Explore in graph

Co-occurring keywords · papers mentioning both

perplexity (4) large language model (4) fine tuning (4) sparsity (3) quantization (3) layer-wise (2) efficiency (2) parameter (2) transfer ability (2) generalization (2)

Papers

Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs “Difficult" Downstream Tasks in LLMs ICML 2024 Non-transferable Pruning ECCV 2024 Optimal Eye Surgeon: Finding image priors through sparse generators at initialization ICML 2024 Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity ICML 2024 SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks ICML 2024 Soft Prompt Recovers Compressed LLMs, Transferably ICML 2024 SPADE: Sparsity-Guided Debugging for Deep Neural Networks ICML 2024 Sparsest Models Elude Pruning: An Exposé of Pruning’s Current Capabilities ICML 2024