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/tsb0601/emp-ssl
공식 코드EMP-SSL: Towards Self-Supervised Learning in One Training Epoch
https://github.com/tsingmaoai/mi-optimize
공식 코드Evaluating the Generalization Ability of Quantized LLMs: Benchmark, Analysis, and Toolbox
https://github.com/ttumyche/mxq-vae
공식 코드Unconditional Image-Text Pair Generation with Multimodal Cross Quantizer
https://github.com/tunglm2203/tirl
공식 코드On the Perturbed States for Transformed Input-robust Reinforcement Learning
https://github.com/tunglm2203/vq_robust_rl
공식 코드Mitigating Adversarial Perturbations for Deep Reinforcement Learning via Vector Quantization
https://github.com/tuskanny/kannolo
공식 코드kANNolo: Sweet and Smooth Approximate k-Nearest Neighbors Search
https://github.com/twilight92z/quantize-watermark
공식 코드https://github.com/two-tiger/fsg
공식 코드Fast and Slow Gradient Approximation for Binary Neural Network Optimization
https://github.com/tyler-hayes/REMIND
공식 코드REMIND Your Neural Network to Prevent Catastrophic Forgetting
https://github.com/tyshiwo/FHR_alignment
공식 코드Towards Highly Accurate and Stable Face Alignment for High-Resolution Videos
https://github.com/uber-research/permute-quantize-finetune
공식 코드Permute, Quantize, and Fine-tune: Efficient Compression of Neural Networks
https://github.com/ubiquitouslearning/mobiledlframeworksbenchmark
공식 코드https://github.com/ucbrise/actnn
공식 코드ActNN: Reducing Training Memory Footprint via 2-Bit Activation Compressed Training
https://github.com/ucdvision/compact3d
공식 코드CompGS: Smaller and Faster Gaussian Splatting with Vector Quantization
https://github.com/uchi-jcl/cachegen
공식 코드CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving
https://github.com/uds-lsv/phone2unit
공식 코드An Information-Theoretic Analysis of Self-supervised Discrete Representations of Speech
https://github.com/ugonfor/DGQ
공식 코드DGQ: Distribution-Aware Group Quantization for Text-to-Image Diffusion Models
https://github.com/uiuc-focal-lab/stochastic-monkeys
공식 코드Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment
https://github.com/ulf1/sentence-embedding-evaluation-german
공식 코드https://github.com/umass-embodied-agi/commvq
공식 코드CommVQ: Commutative Vector Quantization for KV Cache Compression
https://github.com/umich-bipedlab/optimal_shape_generation
공식 코드https://github.com/umich-bipedlab/optimal_shape_global_pose_estimation
공식 코드https://github.com/una-dinosauria/Rayuela.jl
공식 코드https://github.com/unicomai/deepseek-eval
공식 코드Quantitative Analysis of Performance Drop in DeepSeek Model Quantization
https://github.com/unites-lab/moe-quantization
공식 코드Examining Post-Training Quantization for Mixture-of-Experts: A Benchmark
https://github.com/unixpickle/vq-draw
공식 코드https://github.com/upunaprosk/quantized-lm-confidence
공식 코드When Quantization Affects Confidence of Large Language Models?
https://github.com/usamec/double_binary
공식 코드Addition is almost all you need: Compressing neural networks with double binary factorization
https://github.com/ustc-imcl/envc
공식 코드Learning Cross-Scale Weighted Prediction for Efficient Neural Video Compression
https://github.com/ustc-imcl/nvtc
공식 코드https://github.com/ustc-imcl/rdo-gaussian
공식 코드End-to-End Rate-Distortion Optimized 3D Gaussian Representation
https://github.com/ustcivclab/IPM
공식 코드Partition Map-Based Fast Block Partitioning for VVC Inter Coding
https://github.com/ustcwhy/bitvla
공식 코드BitVLA: 1-bit Vision-Language-Action Models for Robotics Manipulation
https://github.com/usyd-fsalab/fp6_llm
공식 코드FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algorithm-System Co-Design
https://github.com/utcsilab/wideband-llr-deep
공식 코드출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.