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/tencent-ailab/muq
공식 코드MuQ: Self-Supervised Music Representation Learning with Mel Residual Vector Quantization
https://github.com/tencentarc/seed-voken
공식 코드Taming Scalable Visual Tokenizer for Autoregressive Image Generation
https://github.com/tencentarc/toklip
공식 코드TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and Generation
https://github.com/tencentarc/vqfr
공식 코드VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder
https://github.com/tencentyouturesearch/crowdcounting-uepnet
공식 코드Uniformity in Heterogeneity:Diving Deep into Count Interval Partition for Crowd Counting
https://github.com/tensorpack/tensorpack/tree/master/examples/DoReFa-Net
공식 코드DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
https://github.com/thakur-nandan/income
공식 코드https://github.com/thanosm97/gsoc2022-openvino
공식 코드Post-training Model Quantization Using GANs for Synthetic Data Generation
https://github.com/the-inscrutable-x/tacq
공식 코드Task-Circuit Quantization: Leveraging Knowledge Localization and Interpretability for Compression
https://github.com/thecommonirin/qresafe
공식 코드https://github.com/thisisbillhe/efficientdm
공식 코드EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models
https://github.com/thomasahle/cce
공식 코드Clustering the Sketch: A Novel Approach to Embedding Table Compression
https://github.com/thu-ml/2by4-pretrain-acc-examples
공식 코드S-STE: Continuous Pruning Function for Efficient 2:4 Sparse Pre-training
https://github.com/thu-ml/Jetfire-INT8Training
공식 코드https://github.com/thu-ml/SageAttention
공식 코드SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration
https://github.com/thu-ml/adaptive-sparse-trainer
공식 코드Pruning Large Language Models with Semi-Structural Adaptive Sparse Training
https://github.com/thu-ml/iodf
공식 코드Fast Lossless Neural Compression with Integer-Only Discrete Flows
https://github.com/thu-ml/tetrajet-mxfp4training
공식 코드https://github.com/thu-nics/mbq
공식 코드MBQ: Modality-Balanced Quantization for Large Vision-Language Models
https://github.com/thu-nics/pm-kvq
공식 코드PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs
https://github.com/thu-nics/qllm-eval
공식 코드https://github.com/thuccslab/mergeguard
공식 코드https://github.com/thudm/glm-130b
공식 코드https://github.com/thuml/DPH
공식 코드https://github.com/thunlp/delta-come
공식 코드Delta-CoMe: Training-Free Delta-Compression with Mixed-Precision for Large Language Models
https://github.com/thupchnsky/mufc
공식 코드https://github.com/thupchnsky/sc_mvae
공식 코드Federated Classification in Hyperbolic Spaces via Secure Aggregation of Convex Hulls
https://github.com/tianzheng4/Distributionally-Adversarial-Attack
공식 코드Towards Sharper First-Order Adversary with Quantized Gradients
https://github.com/timdettmers/bitsandbytes
공식 코드LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
https://github.com/tjrudrnr2/PRISM_ICLR25
공식 코드PRISM: Privacy-Preserving Improved Stochastic Masking for Federated Generative Models
https://github.com/tong-wu-umn/spike-compression-autoencoder
공식 코드Deep Compressive Autoencoder for Action Potential Compression in Large-Scale Neural Recording
https://github.com/tonyzhao-jt/LLM-PQ
공식 코드LLM-PQ: Serving LLM on Heterogeneous Clusters with Phase-Aware Partition and Adaptive Quantization
https://github.com/transformer-vq/transformer_vq
공식 코드Transformer-VQ: Linear-Time Transformers via Vector Quantization
https://github.com/treeai-lab/awesome-kv-cache-management
공식 코드A Survey on Large Language Model Acceleration based on KV Cache Management
https://github.com/trneedham/QuantizedGromovWasserstein
공식 코드출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.