HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Machine Translation On Wmt2016 English German

평가 지표

BLEU score

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	BLEU score	Paper Title	Repository
FLAN 137B (few-shot, k=11)	26.1	Finetuned Language Models Are Zero-Shot Learners
Exploiting Mono at Scale (single)	-	Exploiting Monolingual Data at Scale for Neural Machine Translation	-
FLAN 137B (zero-shot)	27.0	Finetuned Language Models Are Zero-Shot Learners
SMT + iterative backtranslation (unsupervised)	18.23	Unsupervised Statistical Machine Translation
Unsupervised S2S with attention	9.64	Unsupervised Machine Translation Using Monolingual Corpora Only
MADL	40.68	Multi-Agent Dual Learning	-
BiRNN + GCN (Syn + Sem)	24.9	Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks	-
Transformer	26.7	On the adequacy of untuned warmup for adaptive optimization
DeLighT	28.0	DeLighT: Deep and Light-weight Transformer
Attentional encoder-decoder + BPE	34.2	Edinburgh Neural Machine Translation Systems for WMT 16
Unsupervised NMT + weight-sharing	10.86	Unsupervised Neural Machine Translation with Weight Sharing
Linguistic Input Features	28.4	Linguistic Input Features Improve Neural Machine Translation

0 of 12 row(s) selected.