← Optimization & Theory
Deep Learning › Optimization & Theory ›

Model Compression

1674 directly classified papers

Papers per year

Papers

Removing Batch Normalization Boosts Adversarial Training ICML 2022
Optimal Clipping and Magnitude-aware Differentiation for Improved Quantization-aware Training ICML 2022
AdapLeR: Speeding up Inference by Adaptive Length Reduction ACL 2022
Attention Temperature Matters in Abstractive Summarization Distillation ACL 2022
Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm ACL 2022
Multi-Granularity Structural Knowledge Distillation for Language Model Compression ACL 2022
FrugalScore: Learning Cheaper, Lighter and Faster Evaluation Metrics for Automatic Text Generation ACL 2022
Composable Sparse Fine-Tuning for Cross-Lingual Transfer ACL 2022
Probing Structured Pruning on Multilingual Pre-trained Models: Settings, Algorithms, and Efficiency ACL 2022
bert2BERT: Towards Reusable Pretrained Language Models ACL 2022
Robust Lottery Tickets for Pre-trained Language Models ACL 2022
Token Dropping for Efficient BERT Pretraining ACL 2022
Compression of Generative Pre-trained Language Models via Quantization ACL 2022
E-LANG: Energy-Based Joint Inferencing of Super and Swift Language Models ACL 2022
SDR: Efficient Neural Re-ranking using Succinct Document Representation ACL 2022
BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models ACL 2022
TextPruner: A Model Pruning Toolkit for Pre-Trained Language Models ACL 2022
BMInf: An Efficient Toolkit for Big Model Inference and Tuning ACL 2022
Finding the Dominant Winning Ticket in Pre-Trained Language Models ACL 2022
Aligned Weight Regularizers for Pruning Pretrained Neural Networks ACL 2022
Diverse Lottery Tickets Boost Ensemble from a Single Pretrained Model ACL 2022
Parameter-Efficient Abstractive Question Answering over Tables or Text ACL 2022
Knowledge Base Index Compression via Dimensionality and Precision Reduction ACL 2022
Universality of Winning Tickets: A Renormalization Group Perspective ICML 2022
DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale ICML 2022