https://github.com/mit-han-lab/llm-awq
공식 코드AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
Papers with Code 아카이브
Quantization
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 2
- 평가표
- 10
- 보존된 평가 기록
- 32
Quantization is a promising technique to reduce the computation cost of neural network training, which can replace high-cost floating-point numbers (e.g., float32) with low-cost fixed-point numbers (e.g., int8/int16).
Source: Adaptive Precision Training: Quantify Back Propagation in Neural Networks with Fixed-point Numbers
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/mit-han-lab/neurips-micronet
공식 코드https://github.com/mit-han-lab/nunchaku
공식 코드SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models
https://github.com/mit-han-lab/omniserve
공식 코드QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving
https://github.com/mit-han-lab/pytorch-quantum
공식 코드QuantumNAT: Quantum Noise-Aware Training with Noise Injection, Quantization and Normalization
https://github.com/mit-han-lab/qserve
공식 코드QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving
https://github.com/mit-han-lab/smoothquant
공식 코드SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
https://github.com/mit-han-lab/torchquantum
공식 코드QuantumNAT: Quantum Noise-Aware Training with Noise Injection, Quantization and Normalization
https://github.com/ml4its/timevqvae
공식 코드Vector Quantized Time Series Generation with a Bidirectional Prior Model
https://github.com/mlaetvayn/aecct
공식 코드https://github.com/mlcommons/tiny_results_v0.7/tree/main/open/hls4ml-finn
공식 코드https://github.com/mlech26l/quantization_aware_ibp
공식 코드https://github.com/mlzxy/qsparse
공식 코드Training Deep Neural Networks with Joint Quantization and Pruning of Weights and Activations
https://github.com/mlzxy/qsparse-examples
공식 코드Training Deep Neural Networks with Joint Quantization and Pruning of Weights and Activations
https://github.com/mmmings/granular-dq
공식 코드https://github.com/model-compression/lossless_compression
공식 코드"Lossless" Compression of Deep Neural Networks: A High-dimensional Neural Tangent Kernel Approach
https://github.com/modelscope/ms-swift
공식 코드https://github.com/modelscope/swift
공식 코드https://github.com/modeltc/llmc
공식 코드LLMC: Benchmarking Large Language Model Quantization with a Versatile Compression Toolkit
https://github.com/modeltc/mqbench
공식 코드MQBench: Towards Reproducible and Deployable Model Quantization Benchmark
https://github.com/modeltc/outlier_suppression_plus
공식 코드https://github.com/modeltc/qllm
공식 코드QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Models
https://github.com/modeltc/tfmq-dm
공식 코드TFMQ-DM: Temporal Feature Maintenance Quantization for Diffusion Models
https://github.com/moe-inf/awesome-moe-inference
공식 코드A Survey on Inference Optimization Techniques for Mixture of Experts Models
https://github.com/mohaimenz/acdnet
공식 코드https://github.com/mohaimenz/pruning-xnor
공식 코드Pruning vs XNOR-Net: A Comprehensive Study of Deep Learning for Audio Classification on Edge-devices
https://github.com/mohammad-mozaffari/slim
공식 코드SLiM: One-shot Quantization and Sparsity with Low-rank Approximation for LLM Weight Compression
https://github.com/molereddy/long-context-quantization
공식 코드Does quantization affect models' performance on long-context tasks?
https://github.com/moranshkolnik/RobustQuantization
공식 코드https://github.com/mosesananta/qless
공식 코드QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning
https://github.com/mostafaelhoushi/DeepShift
공식 코드https://github.com/motefly/LMA
공식 코드https://github.com/mrusci/training-mixed-precision-quantized-networks
공식 코드https://github.com/mzchen0/UVeQFed
공식 코드UVeQFed: Universal Vector Quantization for Federated Learning
https://github.com/mzhang367/opqn
공식 코드Orthonormal Product Quantization Network for Scalable Face Image Retrieval
https://github.com/nadavlab/CPLLM
공식 코드출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.