Guide-yakabatana mibvunzo · Pakati Nhanho

WordPiece Tokenization Quiz

Ongorora mazwi eWordPiece uye mitemo yemakaro yezvikamu, kusanganisira basa rayo mumitauro yakawanda BERT uye kusiyana kwayo neBPE.

Nzira dzekutungamira dzinoenderanaWordpiece Tokenization
Mubvunzo 1 ye 8

Chii chinonzi '##' prefix chinoratidza muWordPiece kubuda?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Chii chinonzi '##' prefix chinoratidza muWordPiece kubuda?
    • Chiratidzo chidimbu chekuenderera mukati mezwi, kwete izwi-kutanga
    • Chiratidzo chakasarudzika chakasarudzika senge [SEP]
    • Chiratidzo chine huwandu hwepamusoro mumashoko
    • Chiratidzo chiratidzo chisingazivikanwe kunze kwemazwi
  2. Anenge akakura sei izwi rezwi reWordPiece rinoshandiswa neBERT yekutanga?
    • Zvinenge 256 tokeni
    • Zvinenge 30,000 zviratidzo
    • Inenge miriyoni tokeni
    • Anenge 500 tokens
  3. Ko WordPiece's merge criterion inosiyana sei neyakajairwa Byte-Pair Encoding?
    • Iyo inosanganisa iyo isingawanzo pair kuderedza mazwi
    • Haimbosanganisi uye inoshandisa mavara chete
    • Inosanganisirwa nekuwedzera kudzidziswa-dhata mukana pane kuita mbishi frequency
    • Inobatanidza vaviri vaviri panguva yekudzidziswa
  4. What does a BERT-style WordPiece tokenizer return if it cannot segment a word into vocabulary pieces?
    • It always falls back to individual Unicode characters
    • A random subword selected from the vocabulary
    • The original word encoded as one new token
    • The unknown-token marker, such as [UNK]
  5. Panguva yekufungidzira, WordPiece inosarudza sei kupatsanura izwi?
    • Rinopindirana nekazwi kadiki kakarebesa mumashoko nemakaro, kuruboshwe kuenda kurudyi
    • Inotora zvimedu zvipfupi zvinobvira kutanga
    • Inoedza kupatsanurwa kwese kunobvira uye kuaenzanisa
    • Inoshandisa neural network kufanotaura kupatsanurwa
  6. Ndeipi inozivikanwa mhuri yemuenzaniso iyo WordPiece inonyanya kushamwaridzana nayo?
    • GPT mhando
    • BERT uye zvinobva kwayo
    • LLaMA mhando
    • Stable Diffusion
  7. What kind of vocabulary did multilingual BERT use across its training languages?
    • A completely separate vocabulary for every language
    • A vocabulary made only of complete words
    • One shared WordPiece vocabulary
    • A list of byte sequences with no subwords
  8. Which tokenizer class in Hugging Face Transformers constructs a BERT tokenizer that uses WordPiece?
    • GPT2Tokenizer
    • RobertaTokenizer
    • CLIPTokenizer
    • BertTokenizer