AICUP2023-事實文字檢索與查核競賽-前測問卷

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

This questionnaire is a small test for participants before the AICUP competition held by IKM lab.

Since 2018, what has been the most widely used pre-trained model in natural language processing (NLP)?
*

20 points

ELMo

BERT

Cookie Monster

Oscar

Following the last question, what's the architecture used in the model? *

20 points

LSTM (Long short-term memory)

GRU (Gated Recurrent Unit)

VAE (Variational Auto Encoder)

Transformer

What do GPT-2 and GPT-3 mainly use for pre-training?
*

20 points

Masked language modeling

Next-sentence prediction

Language modeling (autoregressive)

Image captioning

What is the main purpose of attention mechanisms?
*

20 points

To allow the model to focus on specific parts of the input sequence

To reduce the computational complexity of the model

To prevent overfitting during training

To reduce the amount of labeled data required for training

What is a solution for retrieving relevant documents? *

20 points

Keyword matching

Topic modeling

Semantic similarity

All of above

Submit

Clear form

This form was created inside of National Cheng Kung University. Report Abuse