清华NLP组年度巨献：机器翻译30年最重要论文阅读清单（上）( 四 ) 新智元报道来源：Github【新智

基于开放词汇和字符的神经机器翻译
FelixHill,KyunghyunCho,SebastienJean,ColineDevin,andYoshuaBengio.2015.EmbeddingWordSimilaritywithNeuralMachineTranslation.InProceedingsofICLR2015.ThangLuong,IlyaSutskever,QuocLe,OriolVinyals,andWojciechZaremba.2015.AddressingtheRareWordProbleminNeuralMachineTranslation.InProceedingsofACL2015.SébastienJean,KyunghyunCho,RolandMemisevic,andYoshuaBengio.2015.OnUsingVeryLargeTargetVocabularyforNeuralMachineTranslation.InProceedingsofACL2015.RicoSennrich,BarryHaddow,andAlexandraBirch.2016.NeuralMachineTranslationofRareWordswithSubwordUnits.InProceedingsofACL2016.Minh-ThangLuongandChristopherD.Manning.2016.AchievingOpenVocabularyNeuralMachineTranslationwithHybridWord-CharacterModels.InProceedingsofACL2016.JunyoungChung,KyunghyunCho,andYoshuaBengio.2016.ACharacter-levelDecoderwithoutExplicitSegmentationforNeuralMachineTranslation.InProceedingsofACL2016.JasonLee,KyunghyunCho,andThomasHofmann.2017.FullyCharacter-LevelNeuralMachineTranslationwithoutExplicitSegmentation.TransactionsoftheAssociationforComputationalLinguistics.YangFeng,ShiyueZhang,AndiZhang,DongWang,andAndrewAbel.2017.Memory-augmentedNeuralMachineTranslation.InProceedingsofEMNLP2017.BaosongYang,DerekF.Wong,TongXiao,LidiaS.Chao,andJingboZhu.2017.TowardsBidirectionalHierarchicalRepresentationsforAttention-basedNeuralMachineTranslation.InProceedingsofEMNLP2017.PeymanPassban,QunLiu,andAndyWay.2018.ImprovingCharacter-BasedDecodingUsingTarget-SideMorphologicalInformationforNeuralMachineTranslation.InProceedingsofNAACL2018.HuadongChen,ShujianHuang,DavidChiang,XinyuDai,andJiajunChen.2018.CombiningCharacterandWordInformationinNeuralMachineTranslationUsingaMulti-LevelAttention.InProceedingsofNAACL2018.FrederickLiu,HanLu,andGrahamNeubig.2018.HandlingHomographsinNeuralMachineTranslation.InProceedingsofNAACL2018.TakuKudo.2018.SubwordRegularization:ImprovingNeuralNetworkTranslationModelswithMultipleSubwordCandidates.InProceedingsofACL2018.MakotoMorishita,JunSuzuki,andMasaakiNagata.2018.ImprovingNeuralMachineTranslationbyIncorporatingHierarchicalSubwordFeatures.InProceedingsofCOLING2018.YangZhao,JiajunZhang,ZhongjunHe,ChengqingZong,andHuaWu.2018.AddressingTroublesomeWordsinNeuralMachineTranslation.InProceedingsofEMNLP2018.ColinCherry,GeorgeFoster,AnkurBapna,OrhanFirat,andWolfgangMacherey.2018.RevisitingCharacter-BasedNeuralMachineTranslationwithCapacityandCompression.InProceedingsofEMNLP2018.RebeccaKnowlesandPhilippKoehn.2018.ContextandCopyinginNeuralMachineTranslation.InProceedingsofEMNLP2018.
训练目标与框架
Marc'AurelioRanzato,SumitChopra,MichaelAuli,andWojciechZaremba.2016.SequenceLevelTrainingwithRecurrentNeuralNetworks.InProceedingsofICLR2016.Minh-ThangLuong,QuocV.Le,IlyaSutskever,OriolVinyals,andLukaszKaiser.2016.Multi-taskSequencetoSequenceLearning.InProceedingsofICLR2016.ShiqiShen,YongCheng,ZhongjunHe,WeiHe,HuaWu,MaosongSun,andYangLiu.2016.MinimumRiskTrainingforNeuralMachineTranslation.InProceedingsofACL2016.SamWisemanandAlexanderM.Rush.2016.Sequence-to-SequenceLearningasBeam-SearchOptimization.InProceedingsofEMNLP2016.DiHe,YingceXia,TaoQin,LiweiWang,NenghaiYu,Tie-YanLiu,Wei-YingMa.2016.DualLearningforMachineTranslation.InProceedingsofNIPS2016.DzmitryBahdanau,PhilemonBrakel,KelvinXu,AnirudhGoyal,RyanLowe,JoellePineau,AaronCourville,andYoshuaBengio.2017.AnActor-CriticAlgorithmforSequencePrediction.InProceedingsofICLR2017.KhanhNguyen,HalDauméIII,andJordanBoyd-Graber.2017.ReinforcementLearningforBanditNeuralMachineTranslationwithSimulatedHumanFeedback.InProceedingsofEMNLP2017.NimaPourdamghaniandKevinKnight.2017.DecipheringRelatedLanguages.InProceedingsofEMNLP2017.DiHe,HanqingLu,YingceXia,TaoQin,LiweiWang,andTieyanLiu.2017.DecodingwithValueNetworksforNeuralMachineTranslation.InProceedingsofNIPS2017.SergeyEdunov,MyleOtt,MichaelAuli,DavidGrangier,andMarc’AurelioRanzato.2018.ClassicalStructuredPredictionLossesforSequencetoSequenceLearning.InProceedingsofNAACL2018.ZhenYang,WeiChen,FengWang,andBoXu.2018.ImprovingNeuralMachineTranslationwithConditionalSequenceGenerativeAdversarialNets.InProceedingsofNAACL2018.PavelPetrushkov,ShahramKhadiviandEvgenyMatusov.2018.LearningfromChunk-basedFeedbackinNeuralMachineTranslation.InProceedingsofACL2018.LijunWu,FeiTian,TaoQin,JianhuangLai,andTie-YanLiu.2018.AStudyofReinforcementLearningforNeuralMachineTranslation.InProceedingsofEMNLP2018.JiataoGu,YongWang,YunChen,KyunghyunCho,andVictorO.K.Li.2018.Meta-LearningforLow-ResourceNeuralMachineTranslation.InProceedingsofEMNLP2018.HarshilShahandDavidBarber.2018.GenerativeNeuralMachineTranslation.InProceedingsofNeurIPS2018.LijunWu,FeiTian,YingceXia,YangFan,TaoQin,JianhuangLai,andTie-YanLiu.2018.LearningtoTeachwithDynamicLossFunctions.InProceedingsofNeurIPS2018.YirenWang,YingceXia,TianyuHe,FeiTian,TaoQin,ChengXiangZhai,andTie-YanLiu.2019.Multi-AgentDualLearning.InProceedingsofICLR2019.