Ọkọọkan-si-tẹle Models adanwo

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

1. Kini awọn paati akọkọ meji ti awoṣe lẹsẹsẹ-si-tẹle?
   • Ohun kooduopo ati ki o kan decoder
   • A tokenizer ati ki o kan classifier
   • Meji aami encoders
   • A àlẹmọ ati ki o kan ranker
2. Iṣoro bọtini wo ni ẹrọ akiyesi yanju ni awọn awoṣe seq2seq?
   • O lọra tokenization
   • Igo igo fekito gigun-ipari fun awọn igbewọle gigun
   • Aini ti fokabulari
   • Awọn data ikẹkọ ti o pọju
3. Kini faaji ti atilẹba 2014 seq2seq awoṣe itumọ lo?
   • Convolutional nẹtiwọki
   • Awọn igi ipinnu
   • Awọn LSTM tolera
   • Igboro ifunni-siwaju fẹlẹfẹlẹ
4. Bawo ni ifarabalẹ ṣe kọ ọrọ-ọrọ fun igbesẹ iyipada kọọkan?
   • Nipa aropin gbogbo awọn ami igbewọle ni dọgbadọgba
   • Nipa aibikita kooduopo patapata
   • Nipa didakọ tokini igbewọle to kẹhin
   • Nipa iye iwọn ti awọn ipinlẹ kooduopo nipa lilo awọn ikun titete
5. Kini idi ti awọn abajade seq2seq le yato ni gigun lati awọn igbewọle?
   • Decoder n ṣe awọn ami ami jade titi ti o fi njade ifihan agbara iduro
   • Awọn igbewọle ati awọn igbejade nigbagbogbo jẹ ipari dogba
   • Awọn paadi kooduopo ohun gbogbo
   • Gigun jẹ ti o wa titi ilosiwaju
6. Awọn awoṣe ode oni wo ni awọn faaji seq2seq ti o da lori Transformer olokiki?
   • ResNet ati VGG
   • T5 ati BART
   • K-tumosi ati PCA
   • Word2Vec ati GloVe
7. In encoder-decoder attention, which representations are commonly used as keys and values?
   • Only the decoder’s future hidden states
   • The vocabulary file’s byte offsets
   • The encoder’s output representations
   • The optimizer’s learning-rate history
8. What does a bidirectional encoder provide to an attention-based decoder?
   • A fixed one-token output regardless of the input
   • A guarantee that output length equals input length
   • A causal mask over future target tokens
   • Contextual representations of source tokens that can use information from both directions