https://github.com/gnobitab/RectifiedFlow
공식 코드Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
Discriminative Region Proposal Adversarial Networks for High-Quality Image-to-Image Translation
LBM: Latent Bridge Matching for Fast Image-to-Image Translation
It's not Greek to mBERT: Inducing Word-Level Translations from Multilingual BERT
CVSS Corpus and Massively Multilingual Speech-to-Speech Translation
TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languages
Consistency by Agreement in Zero-shot Neural Machine Translation
MaskSketch: Unpaired Structure-guided Masked Image Generation
MetricX-24: The Google Submission to the WMT 2024 Metrics Shared Task
mT5: A massively multilingual pre-trained text-to-text transformer
Bilex Rx: Lexical Data Augmentation for Massively Multilingual Machine Translation
Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates
Massive Exploration of Neural Machine Translation Architectures
Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation
Sequence-to-sequence neural network models for transliteration
Attention is Not Only a Weight: Analyzing Transformers with Vector Norms
Rawformer: Unpaired Raw-to-Raw Translation for Learnable Camera ISPs
BilBOWA: Fast Bilingual Distributed Representations without Word Alignments
Bi-SimCut: A Simple Strategy for Boosting Neural Machine Translation
Towards Boosting Many-to-Many Multilingual Machine Translation with Large Language Models
Learning Multilingual Sentence Representations with Cross-lingual Consistency Regularization
An Empirical Study of Consistency Regularization for End-to-End Speech-to-Text Translation
Variational Interaction Information Maximization for Cross-domain Disentanglement
Approaching Neural Grammatical Error Correction as a Low-Resource Machine Translation Task
CommitBERT: Commit Message Generation Using Pre-Trained Programming Language Model
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.