Six Layers Less: Encoder Pruning for Whisper with Label-Free Recovery
Pruning LLMs Like a Physicist: Block Removal as an Ising Optimization Problem
Neural Spectral Capacity: Measuring and Designing Architectures from Network Specification Alone
Higher-order pruning of experts in mixture-of-experts language models
1
What Breaks Under Pruning in Smart Homes, and When? Evaluating LLM Degradation Across Architectures and Task Complexity
SQS: Bayesian DNN Compression through Sparse Quantized Sub-distributions
