Seminar > Paper Reviews

논문 제목

BERT-DST: Scalable End-to-End Dialogue State Tracking with Bidirectional Encoder Representations from Transformer 

Improving BERT Fine-Tuning via Self-Ensemble and Self-Distillation 

Efficient Estimation of Word Representations in Vector Space

Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference

SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient  

WIKIBIAS: Detecting Multi-Span Subjective Biases in Language 

Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context

Probing Commonsense Explanation in Dialogue Response Generation

Personal Research - ESG

Enriching Word Vectors with Subword Information

AEDA: An Easier Data Augmentation Technique for Text Classification

AutoRec: Autoencoders Meet Collaborative Flitering

XLNet: Generalized Autoregressive Pretraining for Language Understanding 

A Simple Language Model for Task-Orieted Dialogue

Unsupervised Data Augmentation for Consistency Training

A Flexible Multi-Task Model for BERT Serving

KorNLI and KorSTS: New Benchmark Datasets for Korean Natural Natural Language Understanding

Neural Graph Collaborative Filtering

Jointly Optimizing State Operation Prediction and Value Generation for Dialogue State Tracking