Translation-Service

Wani sabon tsarin NMT mai haɓaka da bincike wanda ke amfani da bincike mai kwatankwaci da ɓoyayyen matakai don yin amfani da tunawa da fassara iri-iri, marasa maimaitawa don inganta ingancin fassara.

Analysis of degeneracies in inverse propensity scoring for offline machine translation learning from deterministic logs, with proposed solutions and formal insights.

Nazari mai kwatanta ƙamus na bugu/na kan layi da tushen kalmomi, mai mai da hankali kan ci gabansu, amincinsu, da makomarsu a fasahar fassarar.

Binciken aikace-aikacen farko na Fassarar Injin Neural akan Larabci, kwatanta shi da tsarin tushen jumla, binciken tasirin sarrafa kafin aiwatarwa, da kimanta ƙarfin jurewa ga sauyin yanki.

Gabatar da WOKIE, tsarin buɗaɗɗen kayan aiki don fassarar atomatik na tsarin ma'ana na SKOS ta amfani da sabis na fassara na waje da inganta LLM don haɓaka samun dama da haɗin kai tsakanin harsuna a Nazarin Al'adu na Dijital.

Bincike kan haɓaka fassarar inji ta tushen LLM ta amfani da ƙwaƙwalwar fassara, yana samun gagarumin ci gaba a cikin maki BLEU a cikin harsuna da yanki daban-daban.

Bincike kan samfuran koyo mai zurfi masu sauƙi, na gida don fassara harsuna masu ƙarancin albarkatu, da aka ɓoye, da na musamman ta amfani da tsarin LSTM-RNN encoder-decoder.

Nazarin tsarin Fassarar Injini da aka ƙera don harsunan Indiya, ya ƙunshi hanyoyi kamar Kai tsaye, Na Ƙa'ida, da Na Tushen Bayanai, manyan tsare-tsare, da hanyoyin bincike na gaba.

Binciken amfani da LLMs don fassarar bayanan fassarar ma'ana na Turanci zuwa harsuna daban-daban don horar da masu fassara, wanda ya fi kyau fiye da hanyoyin fassara-turawa a cikin harsuna 50.

Bincike kan DGT-TM, ƙwaƙwalwar fassara da ake samu kyauta wacce ta ƙunshi harsuna 22 na EU da nau'ikan fassara 231, yadda aka ƙirƙira ta, aikace-aikacenta a fasahar harshe, da tasirinta na gaba.

Nazarin takarda da ke gabatar da sabon tsarin Koyon Ƙarfafawa na Mai Ba da Shawara da Mai Zargi (A2C) don fassarar inji mai nau'i daban-daban, wanda ya haɗa bayanan gani da na rubutu.

Analysis of a hybrid NMT-SMT framework that integrates SMT recommendations into NMT decoding to address fluency-adequacy trade-offs, with experimental results on Chinese-English translation.

Cikakken bincike kan Fassarar Injin Neuronal (NMT) wanda ya ƙunshi tarihinta, ainihin ra'ayoyin hanyoyin sadarwa na neuronal, tsarin masu shigar da bayanai da masu fitarwa, gyare-gyare, da ƙalubalen yanzu.

Tsarin zurfin koyo mai haɗa kai da kai wanda ya haɗa kima da inganci da gyara ta atomatik don inganta sakamakon fassarar inji da rage nauyin aikin mai fassara na ɗan adam.

Binciken TM-LevT, wani sabon nau'i na Transformer na Levenshtein da aka tsara don gyara fassarori daga ƙwaƙwalwar fassara yadda ya kamata, inda ya kai ga aikin daidai da samfuran da suka dogara ga baya.

Bincika hanyoyin lissafi na ayyuka masu ƙarancin ƙarfi don zaɓin mafi kyawun misali a cikin fassarar injin jijiyoyi da aka ƙarfafa da ma'ajiya, tare da mai da hankali kan inganta ɗaukar hoto.

Bincike kan sabuwar hanyar daidaita yanki bayan horarwa don Fassarar Injin Jijiya (NMT), tare da bincika ƙwarewa ta hankali, sakamakon gwaji, da aikace-aikace na gaba.

Nazarin kalubale da damammaki a cikin horar da masu fassara shari'a, tare da bincika sabbin tsare-tsare, haɗa fasaha, da kuma sauyin matsayin masu fassara a matsayin masu sulhunta al'adu.

Analysis of global text augmentation methods (Word2Vec, WordNet, round-trip translation) and mixup for improving short text classification performance and model robustness.

Ya gabatar da Gwajin Tsarin Tsarin (SIT), wata hanyar gwaji ta metamorphic don tabbatar da software na fassarar injin ta hanyar nazarin daidaiton tsari a cikin sakamakon fassara.

Nazarin yadda ka'idojin da'a na ƙwararrun fassara suke aiki a fagen da ba na ƙwararru ba kamar taron jama'a, fassarar masoya, da kuma sa harshe na FOSS, tare da bayyana bambance-bambance da sabbin abubuwa.

Nazarin algorithms na kamanceceniya don tsarin Ƙwaƙwalwar Fassara, kimanta alaƙarsu da hukunce-hukuncen mutane da kuma gabatar da sabuwar hanyar daidaitaccen ma'auni na n-gram mai nauyi.

Bincike kan NMT da aka ƙarfafa da ƙwaƙwalwar fassara ta mahangar yiwuwa da rarraba bambanci da karkacewa, bayanin sabani na aiki da gabatar da ingantacciyar hanyar haɗaɗɗun tsari.

Bincike kan kayan aikin QA na zamani don fassara, haɗuwa da kayan aikin CAT, ma'auni na masana'antu, da tantance sakamakon software na QA masu zaman kansu.

Bincike kan yadda duniya baki daya da fasaha ke sake fasalin bukatun fassara, inda ake sanya masu fassara a matsayin masu shiga tsakani na al'adu da kayan aiki na dabarun kasuwanci.

Analyzes a variational encoder-decoder model for neural machine translation, which introduces continuous latent variables to explicitly model underlying semantics, thereby improving translation quality.

Bincike akan SM2, tsarin Transformer Transducer mai gudana don ASR da fassarar magana na yare da yawa, mai siffar haƙiƙanin ikon sifili-sifili da kulawa mara ƙarfi.