Papers with Code 아카이브

Model Compression

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법 전체 데이터셋

하위 태스크: 1
평가표: 2
보존된 평가 기록: 14

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

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/hanguo97/lq-lora

공식 코드

LQ-LoRA: Low-rank Plus Quantized Matrix Decomposition for Efficient Language Model Finetuning

https://github.com/hankye/pagcp

공식 코드

Performance-aware Approximation of Global Channel Pruning for Multitask CNNs

https://github.com/haochen-rye/nerv

공식 코드

NeRV: Neural Representations for Videos

https://github.com/haolibai/Cross-Distillation

공식 코드

Few Shot Network Compression via Cross Distillation

https://github.com/haria/SqueezeNet

공식 코드

SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

https://github.com/hartvigsen-group/composable-interventions

공식 코드

Composable Interventions for Language Models

https://github.com/harveyp123/iccd_sptrn_slr

공식 코드

Towards Sparsification of Graph Neural Networks

https://github.com/hatchetProject/QuEST

공식 코드

QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning

https://github.com/hchoi71/mix-kd

공식 코드

Understanding the Role of Mixup in Knowledge Distillation: An Empirical Study

https://github.com/hdzhang-code/smallID

공식 코드

Minimizing PLM-Based Few-Shot Intent Detectors

https://github.com/hftsoi/symbolnet

공식 코드

SymbolNet: Neural Symbolic Regression with Adaptive Dynamic Pruning for Compression

https://github.com/hkuds/lightgnn

공식 코드

LightGNN: Simple Graph Neural Network for Recommendation

https://github.com/hkuds/promptmm

공식 코드

PromptMM: Multi-Modal Knowledge Distillation for Recommendation with Prompt-Tuning

https://github.com/hly1998/brcd

공식 코드

Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing

https://github.com/holzbock/dfbf

공식 코드

Data-Free Backbone Fine-Tuning for Pruned Neural Networks

https://github.com/hrcheng1066/awesome-pruning

공식 코드

A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendations

https://github.com/hsndbk4/gdws

공식 코드

Generalized Depthwise-Separable Convolutions for Adversarially Robust and Efficient Neural Networks

https://github.com/htqin/IR-Net

공식 코드

Forward and Backward Information Retention for Accurate Binary Neural Networks

https://github.com/huangowen/qat-acs

공식 코드

Efficient and Robust Quantization-aware Training via Adaptive Coreset Selection

https://github.com/huangowen/quantization-variation

공식 코드

Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision

https://github.com/huangowen/vvtq

공식 코드

Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision

https://github.com/huawei-noah/Data-Efficient-Model-Compression/tree/master/DAFL

공식 코드

Data-Free Learning of Student Networks

https://github.com/huawei-noah/Efficient-Computing

공식 코드

Towards Higher Ranks via Adversarial Weight Pruning

https://github.com/huawei-noah/Pretrained-Language-Model

공식 코드

BinaryBERT: Pushing the Limit of BERT Quantization

https://github.com/huawei-noah/vega

공식 코드

VEGA: Towards an End-to-End Configurable AutoML Pipeline

https://github.com/hui-po-wang/LM-GC

공식 코드

Language Models as Zero-shot Lossless Gradient Compressors: Towards General Neural Parameter Prior Models

https://github.com/hustvl/pd-quant

공식 코드

PD-Quant: Post-Training Quantization based on Prediction Difference Metric

https://github.com/idstcv/kda

공식 코드

Improved Knowledge Distillation via Full Kernel Matrix Transfer

https://github.com/ifding/MLKD

공식 코드

Multi-level Knowledge Distillation via Knowledge Alignment and Correlation

https://github.com/ijcai2023/uni_kd

공식 코드

Distilling Universal and Joint Knowledge for Cross-Domain Model Compression on Time Series Data

https://github.com/imrl/gsb-vision-transformer

공식 코드

GSB: Group Superposition Binarization for Vision Transformer with Limited Training Samples

https://github.com/intel/intel-extension-for-transformers

공식 코드

Fast DistilBERT on CPUs

https://github.com/intellabs/hardware-aware-automated-machine-learning

공식 코드

MultiPruner: Balanced Structure Removal in Foundation Models

https://github.com/intersun/CoDIR

공식 코드

Contrastive Distillation on Intermediate Representations for Language Model Compression

https://github.com/intersun/PKD-for-BERT-Model-Compression

공식 코드

Patient Knowledge Distillation for BERT Model Compression

https://github.com/ist-daslab/evopress

공식 코드

EvoPress: Towards Optimal Dynamic Model Compression via Evolutionary Search

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.