Model Compression

Model Compression is an actively pursued area of research over the last few years with the goal of deploying state-of-the-art deep networks in low-power and resource limited devices without significant drop in accuracy. Parameter pruning, low-rank factorization and weight quantization are some of the proposed methods to compress the size of deep networks.

Source: KD-MRI: A knowledge distillation framework for image reconstruction and image restoration in MRI workflow

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 281–290 of 1356 papers

Title	Date	Tasks	Status	Score
How does topology of neural architectures impact gradient propagation and model performance?	Jun 16, 2021	Model CompressionNeural Architecture Search	CodeCode Available	5
Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning	Nov 19, 2015	Atari GamesDeep Reinforcement Learning	CodeCode Available	5
HRKD: Hierarchical Relational Knowledge Distillation for Cross-domain Language Model Compression	Oct 16, 2021	Few-Shot LearningKnowledge Distillation	CodeCode Available	5
ImPart: Importance-Aware Delta-Sparsification for Improved Model Compression and Merging in LLMs	Apr 17, 2025	Model CompressionQuantization	CodeCode Available	5
Bayesian Tensorized Neural Networks with Automatic Rank Selection	May 24, 2019	Model CompressionTensor Decomposition	CodeCode Available	5
Gradual Channel Pruning while Training using Feature Relevance Scores for Convolutional Neural Networks	Feb 23, 2020	Model Compression	CodeCode Available	5
GSB: Group Superposition Binarization for Vision Transformer with Limited Training Samples	May 13, 2023	BinarizationKnowledge Distillation	CodeCode Available	5
Bayesian Optimization with Clustering and Rollback for CNN Auto Pruning	Sep 22, 2021	Bayesian OptimizationClustering	CodeCode Available	5
Generalizing Teacher Networks for Effective Knowledge Distillation Across Student Architectures	Jul 22, 2024	Knowledge DistillationModel Compression	CodeCode Available	5
A Miniaturized Semantic Segmentation Method for Remote Sensing Image	Oct 27, 2018	Model CompressionSemantic Segmentation	CodeCode Available	5

Show:10 25 50

← PrevPage 29 of 136Next →

All datasets ImageNet QNLI

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	ADLIK-MO-ResNet50+W4A4	Top-1	77.88	—	Unverified
2	ADLIK-MO-ResNet50+W3A4	Top-1	77.34	—	Unverified
3	ResNet-18 + 4bit-1dim model compression using DKM	Top-1	70.52	—	Unverified
4	MobileNet-v1 + 4bit-1dim model compression using DKM	Top-1	69.63	—	Unverified
5	ResNet-18 + 2bit-1dim model compression using DKM	Top-1	68.63	—	Unverified
6	MobileNet-v1 + 2bit-1dim model compression using DKM	Top-1	67.62	—	Unverified
7	ResNet-18 + 4bit-4dim model compression using DKM	Top-1	66.1	—	Unverified
8	ResNet-18 + 2bit-2dim model compression using DKM	Top-1	64.7	—	Unverified
9	MobileNet-v1 + 4bit-4dim model compression using DKM	Top-1	61.4	—	Unverified
10	ResNet-18 + 1bit-1dim model compression using DKM	Top-1	59.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MobileBERT + 2bit-1dim model compression using DKM	Accuracy	82.13	—	Unverified
2	MobileBERT + 1bit-1dim model compression using DKM	Accuracy	63.17	—	Unverified