Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska Isku Xidhka Macnaha guud ee ELMo

Dib u eeg sida qaabka luqadda laba jiho ee ELMo uu u abuuro ereyada xasaasiga ah ee macnaha guud iyo isku darka lakabyada soo noqnoqda ee hawlaha hoose.

Wadooyinka haga ee la xidhiidhaElmo Contextual Embeddings
Su'aal 1 ee 8

Waa maxay farqiga ugu muhiimsan ee u dhexeeya ELMo iyo ku-xidhitaanadii hore sida Word2Vec?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Waa maxay farqiga ugu muhiimsan ee u dhexeeya ELMo iyo ku-xidhitaanadii hore sida Word2Vec?
    • ELMo waxay bixisaa isla kelmada kala duwan oo ku xidhan jumladdeeda
    • ELMo waxay isticmaashaa faleebo yaryar
    • ELMo kaliya waxay ku shaqeysaa kelmado kali ah, ee kuma shaqeyso weedho
    • ELMo uma baahna xog tababar
  2. Waa maxay qaab-dhismeedka neerfaha ee ay ELMO u isticmaasho akhrinta qoraalka?
    • Transformer leh is-fiirsi
    • Moodeelka luqadda LSTM ee laba jiho oo qoto dheer
    • Shabakad muuqaal is-daba-joog ah
    • Shabakad fudud oo gudbiye ah
  3. Sidee ELMo isugu geysaa lakabyada gudaha hawsha hoose?
    • Waxay isticmaashaa kaliya lakabka sare ee ugu dambeeya
    • Waxay celcelis ahaan dhammaan lakabyada si siman u hagaajisaa natiijada
    • Waxay adeegsataa la bartay, isku darka miisaanka oo dhan hawl gaar ah
    • Waxay soo qaadaa hal lakab si aan kala sooc lahayn mar kasta
  4. Maxay ELMo u isticmaashaa sidii unugyadeeda wax gelinta si ay u qabato erayada aan la arkin?
    • Raadinta ereyga oo dhan kaliya
    • Aqoonsiga qaamuuska go'an
    • Erayada-hoosaadka byte-labo-ku-xidhan ee BERT
    • Heerarka CNN matalaad
  5. Qiyaastii goorma ayaa ELMo la soo bandhigay yaase keenay?
    • 2018, by Allen Institute for AI cilmi
    • 2013, waxaa qoray Google kooxda Word2Vec
    • 2020, ee OpenAI
    • 2015, by Facebook AI
  6. Which corpus was used to train the bidirectional language model described in the ELMo paper?
    • The One Billion Word Benchmark
    • The Penn Treebank only
    • A corpus of labeled images
    • The GloVe Common Crawl vectors
  7. What language-model prediction does ELMo's forward LSTM learn?
    • Predicting the previous word from future context
    • Assigning a fixed vector to every document
    • Predicting the next word from the preceding context
    • Predicting a sentence's image resolution
  8. What linguistic information do ELMo's lower and higher LSTM layers tend to capture?
    • Lower layers encode image color, while higher layers encode audio pitch
    • Every layer stores only word frequency and ignores context
    • Higher layers are always limited to punctuation prediction
    • Lower layers tend toward syntax, while higher layers capture more context-dependent semantics