https://github.com/utkarsh-dmx/project-resq
공식 코드ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residuals
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/v18nguye/glad
공식 코드GLAD: Improving Latent Graph Generative Modeling with Simple Quantization
https://github.com/vahe1994/aqlm
공식 코드Extreme Compression of Large Language Models via Additive Quantization
https://github.com/vahe1994/spqr
공식 코드SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
https://github.com/vbdi/casp
공식 코드CASP: Compression of Large Multimodal Models Based on Attention Sparsity
https://github.com/vibe007/Dequantization_Diffusion_Models
공식 코드https://github.com/vineeths96/Gradient-Compression
공식 코드Unbiased Single-scale and Multi-scale Quantizers for Distributed Optimization
https://github.com/vision-agh/lift
공식 코드LiFT: Lightweight, FPGA-tailored 3D object detection based on LiDAR data
https://github.com/vita-group/junk_dna_hypothesis
공식 코드https://github.com/vita-group/llm-kick
공식 코드https://github.com/vitaminzl/hqa-gae
공식 코드Hierarchical Vector Quantized Graph Autoencoder with Annealing-Based Code Selection
https://github.com/vityavitalich/taxollama
공식 코드TaxoLLaMA: WordNet-based Model for Solving Multiple Lexical Semantic Tasks
https://github.com/vivswan/gelureluinterpolation
공식 코드https://github.com/vllm-project/vllm
공식 코드AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
https://github.com/vsingh-group/framequant
공식 코드https://github.com/vvvm23/stoi-vqcpc
공식 코드Non-Intrusive Binaural Speech Intelligibility Prediction from Discrete Latent Representations
https://github.com/vvvm23/sundae-vqgan
공식 코드Megapixel Image Generation with Step-Unrolled Denoising Autoencoders
https://github.com/walkerning/aw_nas/tree/master/examples/research/ftt-nas
공식 코드FTT-NAS: Discovering Fault-Tolerant Convolutional Neural Architecture
https://github.com/wangxb96/awesome-ai-on-the-edge
공식 코드Optimizing Edge AI: A Comprehensive Survey on Data, Model, and System Strategies
https://github.com/wcshin-git/te-vqgan
공식 코드https://github.com/wearepal/visual-word-tokenizer
공식 코드Efficient Online Inference of Vision Transformers by Training-Free Tokenization
https://github.com/weihai-98/a-2q
공식 코드$\rm A^2Q$: Aggregation-Aware Quantization for Graph Neural Networks
https://github.com/weihangliu2024/content_aware_nerf
공식 코드https://github.com/weitianxin/mlp_fusion
공식 코드MLP Fusion: Towards Efficient Fine-tuning of Dense and Mixture-of-Experts Language Models
https://github.com/weixiangxu/sttn
공식 코드https://github.com/wfwf10/feature-based-federated-transfer-learning
공식 코드Feature-based Federated Transfer Learning: Communication Efficiency, Robustness and Privacy
https://github.com/wimh966/QDrop
공식 코드QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization
https://github.com/wimh966/outlier_suppression
공식 코드Outlier Suppression: Pushing the Limit of Low-bit Transformer Language Models
https://github.com/wookyounghan/losslessinr
공식 코드Towards Lossless Implicit Neural Representation via Bit Plane Decomposition
https://github.com/wroth8/bnc
공식 코드On Resource-Efficient Bayesian Network Classifiers and Deep Neural Networks
https://github.com/wujiangxu/slmrec
공식 코드SLMRec: Distilling Large Language Models into Small for Sequential Recommendation
https://github.com/wujx2001/QwT
공식 코드https://github.com/wusize/harmon
공식 코드Harmonizing Visual Representations for Unified Multimodal Understanding and Generation
https://github.com/wyf0912/r2lcm
공식 코드https://github.com/xianggebenben/pdADMM-G
공식 코드Towards Quantized Model Parallelism for Graph-Augmented MLPs Based on Gradient-Free ADMM Framework
https://github.com/xiaobigfeng/pmq-ve
공식 코드PMQ-VE: Progressive Multi-Frame Quantization for Video Enhancement
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.