SOTAVerified

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 AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 251–300 of 1356 papers

TitleStatusHype
Compressed models are NOT miniature versions of large models—0
Mamba-PTQ: Outlier Channels in Recurrent Large Language Models—0
Minimizing PLM-Based Few-Shot Intent DetectorsCode0
Inference Optimization of Foundation Models on AI Accelerators—0
Explicit-NeRF-QA: A Quality Assessment Database for Explicit NeRF Model CompressionCode0
Composable Interventions for Language ModelsCode1
Beyond Perplexity: Multi-dimensional Safety Evaluation of LLM CompressionCode0
Quantizing YOLOv7: A Comprehensive Study—0
AMD: Automatic Multi-step Distillation of Large-scale Vision Models—0
The Impact of Quantization and Pruning on Deep Reinforcement Learning Models—0
MLKD-BERT: Multi-level Knowledge Distillation for Pre-trained Language Models—0
Efficient DNN-Powered Software with Fair Sparse Models—0
FoldGPT: Simple and Effective Large Language Model Compression Scheme—0
MCNC: Manifold Constrained Network Compression—0
Q-DiT: Accurate Post-Training Quantization for Diffusion TransformersCode2
LiteYOLO-ID: A Lightweight Object Detection Network for Insulator Defect DetectionCode1
Exploring compressibility of transformer based text-to-music (TTM) models—0
Speeding Up Image Classifiers with Little Companions—0
Pruning via Merging: Compressing LLMs via Manifold Alignment Based Layer MergingCode1
Reinforced Knowledge Distillation for Time Series RegressionCode0
MoA: Mixture of Sparse Attention for Automatic Large Language Model CompressionCode2
FLoCoRA: Federated learning compression with low-rank adaptationCode0
Failure-Resilient Distributed Inference with Model Compression over Heterogeneous Edge Devices—0
SDQ: Sparse Decomposed Quantization for LLM Inference—0
Finding Task-specific Subnetworks in Multi-task Spoken Language Understanding Model—0
Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead—0
An Empirical Investigation of Matrix Factorization Methods for Pre-trained Transformers—0
Model Adaptation for Time Constrained Embodied Control—0
Knowledge Distillation in Federated Learning: a Survey on Long Lasting Challenges and New Solutions—0
Implicit Neural Representation for Videos Based on Residual Connection—0
EncCluster: Scalable Functional Encryption in Federated Learning through Weight Clustering and Probabilistic Filters—0
PC-LoRA: Low-Rank Adaptation for Progressive Model Compression with Knowledge Distillation—0
MobileAIBench: Benchmarking LLMs and LMMs for On-Device Use Cases—0
DistilDoc: Knowledge Distillation for Visually-Rich Document Applications—0
Examining Post-Training Quantization for Mixture-of-Experts: A BenchmarkCode1
On the social bias of speech self-supervised models—0
Slicing Mutual Information Generalization Bounds for Neural NetworksCode0
Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical PerspectiveCode0
Reweighted Solutions for Weighted Low Rank Approximation—0
Towards Efficient Deep Spiking Neural Networks Construction with Spiking Activity based Pruning—0
Robust Knowledge Distillation Based on Feature Variance Against Backdoored Teacher ModelCode0
LCQ: Low-Rank Codebook based Quantization for Large Language Models—0
Effective Interplay between Sparsity and Quantization: From Theory to Practice—0
Occam Gradient DescentCode0
Dual sparse training framework: inducing activation map sparsity via Transformed 1 regularization—0
subMFL: Compatiple subModel Generation for Federated Learning in Device Heterogenous EnvironmentCode0
ExtremeMETA: High-speed Lightweight Image Segmentation Model by Remodeling Multi-channel Metamaterial Imagers—0
Efficient Model Compression for Hierarchical Federated Learning—0
NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models—0
TinyM^2Net-V3: Memory-Aware Compressed Multimodal Deep Neural Networks for Sustainable Edge Deployment—0
Show:102550
← PrevPage 6 of 28Next →
All datasetsImageNetQNLI

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ADLIK-MO-ResNet50+W4A4Top-177.88—Unverified
2ADLIK-MO-ResNet50+W3A4Top-177.34—Unverified
3ResNet-18 + 4bit-1dim model compression using DKMTop-170.52—Unverified
4MobileNet-v1 + 4bit-1dim model compression using DKMTop-169.63—Unverified
5ResNet-18 + 2bit-1dim model compression using DKMTop-168.63—Unverified
6MobileNet-v1 + 2bit-1dim model compression using DKMTop-167.62—Unverified
7ResNet-18 + 4bit-4dim model compression using DKMTop-166.1—Unverified
8ResNet-18 + 2bit-2dim model compression using DKMTop-164.7—Unverified
9MobileNet-v1 + 4bit-4dim model compression using DKMTop-161.4—Unverified
10ResNet-18 + 1bit-1dim model compression using DKMTop-159.7—Unverified
#ModelMetricClaimedVerifiedStatus
1MobileBERT + 2bit-1dim model compression using DKMAccuracy82.13—Unverified
2MobileBERT + 1bit-1dim model compression using DKMAccuracy63.17—Unverified