본문으로 건너뛰기
Juhyeon's Blog
Search
검색
다크 모드
라이트 모드
탐색기
태그: Architecture
12건의 항목
2026년 6월 04일
Attention Residuals
paper
Architecture
ResidualConnection
DepthAttention
AttnRes
PreNorm
KimiLinear
ScalingLaw
MoE
2026년 6월 04일
Auto-Encoding Variational Bayes
paper
VAE
GenerativeModel
VariationalInference
Architecture
Foundational
Kingma
ICLR
2026년 6월 04일
Efficient Estimation of Word Representations in Vector Space
Word2Vec
WordEmbedding
CBOW
SkipGram
DistributedRepresentation
NLP
RepresentationLearning
ICLR2013
Mikolov
Architecture
2026년 6월 04일
Efficiently Modeling Long Sequences with Structured State Spaces
paper
SSM
StateSpaceModel
S4
HiPPO
LongRangeDependencies
NPLR
CauchyKernel
ICLR2022
Architecture
FoundationalPaper
2026년 6월 04일
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
GRU
LSTM
RNN
SequenceModeling
GatedUnit
Architecture
NeurIPS2014
2026년 6월 04일
Gradient-based learning applied to document recognition
Architecture
CNN
LeNet
DocumentRecognition
MNIST
DeepLearning
Classic
LeCun
ConvolutionalNeuralNetwork
2026년 6월 04일
Hierarchical Text-Conditional Image Generation with CLIP Latents
Architecture
DiffusionModels
TextToImage
CLIP
DALL-E2
unCLIP
GenerativeModels
Multimodal
OpenAI
HierarchicalGeneration
2026년 6월 04일
Hyena Hierarchy - Towards Larger Convolutional Language Models
paper
Architecture
SubQuadratic
LongConvolution
HyenaOperator
AttentionFree
SSM
ICML2023
DataControlledGating
2026년 6월 04일
Mamba - Linear-Time Sequence Modeling with Selective State Spaces
paper
SSM
SelectiveSSM
Mamba
Architecture
LinearTime
SelectionMechanism
HardwareAware
ParallelScan
StateSpaceModel
HiPPO
2026년 6월 04일
Neural Machine Translation by Jointly Learning to Align and Translate
Attention
NMT
Encoder-Decoder
BiRNN
ICLR2015
Bahdanau
SoftAlignment
Seq2Seq
Architecture
DeepLearning
2026년 6월 04일
StripedHyena - Moving Beyond Transformers with Hybrid Signal Processing Models
paper
Architecture
HybridModel
StripedHyena
Hyena
Attention
LongContext
SubQuadratic
TogetherAI
BeyondTransformer
ModelGrafting
2026년 6월 04일
Titans - Learning to Memorize at Test Time
Architecture
LongContext
Attention
NeuralMemory
TestTimeLearning
Titans
Transformer
SSM
MetaLearning
AssociativeMemory