README.md

Jiphyeonjeon Season 2 - Intermediate Group

• 수정 및 업데이트 요청 : 김도은([email protected]), 정환석([email protected]), 진명훈([email protected])
• 중급반 스케쥴: 구글 시트
  
  

Presentations

• 01: Google's Neural Machine Translation

  • Video, Presentation
  • Keywords: NMT, Seq2seq, Wordpiece Model, Parallelism
  • Team: 진명훈(발표), 송지현, 지우석

• 02: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

  • Paper, Video, Slides, Demo Code(tokenizer & mlm)
  • Keywords: BERT, trasnformer, MLM, NSP, Pre-training, Fine-tuning
  • Team: 김유빈, 이승미, 이정섭 - 다같이 발표 :)

• 03: Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation

  • Paper, Video, Slides
  • Keywords: Multilingual, Zeroshot Translation
  • Team: 최석웅(발표), 주완, 진승정

• 04: Task-oriented Dialogue System에 대한 넓고 얕은 지식들

  • Video, Slides
  • Keywords: Task-oriented Dialogue System, NLU, Dialogue Manager, NLG, Knowledge Graph
  • Team: 김도은(발표), 금빛나(발표), 김종우

• 05: Cross-lingual Language Model Pretraining

  • Paper, Video, Slides
  • Keywords: Cross-lingual, Pretraining, TLM, XNLI, Fine-tuning
  • Team: 서덕진(발표), 정환석 , 최원석

• 06: Semi-Supervised Sequence Modeling With Cross-View Training

  • Paper, Video, Slides
  • Keywords: Cross-view training, CVT, semi-supervised, Sequence Modeling
  • Team: 김태형(발표), 양재우, 이다영

• 07: What You Can Cram Into a Single $&!#* Vector: Probing Sentence Embeddings For Linguistc Properties

  • Paper, Video, Slides
  • Keywords: Probing Tasks, Sentence Embeddings
  • Team: 이성인(발표), 박석종, 소현지

• 08: Improving Language Understanding by Generative Pre-Training & Language Models are Unsupervised Multitask Learners

  • Paper1, Papaer2, Video, Slides
  • Keywords: GPT1, GPT2
  • Team: 임진호(발표), 허치영(발표)

• 09: A Persona-Based Neural Conversation Model

  • Paper, Video, Slides
  • Keywords: Persona, Open domain conversation
  • Team: 이예진(발표), 구영은(발표), 전하훈

• 10: ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators

  • Paper, Video, Slides
  • Keywords: ELECTRA, small MLM, Generator, Discriminator
  • Team: 고현지, 김병진, 차지윤 (모두 발표)

• 11: Neural document summarization by jointly learning to score and select sentences

  • Paper, Video, Slides
  • Keywords: NEUSUM, Extractive Summarization, document encoder, sentence extractor
  • Team: 김사무엘(발표), 박새란, 이승언

• 12: Adversarial Examples for Evaluating Reading Comprehension Systems

  • Paper, Video, [Slides](https://github.com/jiphyeonjeon/season2/blob/main/intermediate/presentations/집현전_중급_12조_Adversarial Examples.pdf)
  • Keywords: Adversarial􀀁Sentences, Adversarial􀀁Examples, Adversarial􀀁evaluation, SQuAD, BiDAF, Match-LSTM
  • Team: 박병학(발표), 김주찬, 이창호

• 14: Unsupervised Machine Translation Using Monolingual Corpora Only

  • Paper, Video, Slides
  • Keywords: Adversarial Training, Unsupervised Machine Trainslation, AutoEncoding, Seq2Seq
  • Team: 이원호(발표), 김택현, 양수영

• 15: RoBERTa: A Robustly Optimized BERT Pretraining Approach

  • Paper, Video, Slides
  • Keywords: BERT, RoBERTa, transformer, Pre-training, Fine-tuning
  • Team: 김영재, 김재희, 왕대성 - 전원 발표

• 16: A Survey on Deep Learning for Named Entity Recognition

  • Paper, Video, Slides
  • Keywords: NER, CRF, LSTM, Neural Network, Encoder, Decoder
  • Team: 류명현(발표), 이수민(발표), 이아름

• 18: Graph Attention Networks(GAT)

  • Paper, Video, Slides
  • Keywords: Attention, Graph
  • Team: 김남균(발표), 김명환, 김지성

• 19: Universal Sentence Encoder

  • Paper, Video, Slides
  • Keywords: Sentence Embedding, Transfer Learning
  • Team: 김진환(발표), 우연수(발표), 하상천(발표), 하성진(발표)

• 20: Text Summarization with Pretrained Encoders

  • Paper, Video, Slides
  • Keywords: Text summarization, Bertsum, Bert
  • Team: 강민지(발표), 강민구(발표)