Reading group

Introduction

The Natural Language Processing (NLP) reading group meets regularly (usually weekly) to discuss research topics on a particular sub-field of NLP.

Subscription

You can receive announcements about the reading group by joining our mailing list. To join the mailing list, please use an academic email address and send an email to ntunlpsg@gmail.com with a subject “subscribe ntu-nlp”. If you use a non-academic email address, we would have to verify it which could delay your subscription process.

Location

NTU MICL Lab (former MML Lab) meeting room. Room No: N4-B1c-17b

Schedules


2020 Dec | Text Generation and Machine Translation
Time: Friday 15:30 - 17:00

Date Topic Content Presenter(s) Time
11-12-20 Text Generation by Learning from Off-Policy Demonstrations Paper Slide Han Cheol 15:30 - 17:00
(UTC +8)
18-12-20 Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Paper Slide Han Simeng 15:30 - 17:00
(UTC +8)

2020 Aug-Nov | Text Generation, Reinforcement Learning and Others
Time: Wednesday 15:30 - 17:00

Date Topic Content Presenter(s) Time
07-08-20 Beyond Accuracy: Behavioral Testing of NLP Models with CHECKLIST Paper,
Generative-Discriminative Text Classification with Neural Networks Paper
Slide,
Slide
Jwala,
Jwala
15:30 - 17:00
(UTC +8)
21-08-20 Dense Passage Retrieval for Open-Domain Question Answering Paper Slide Megh 15:30 - 17:00
(UTC +8)
28-08-20 Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Paper Slide Han Simeng 15:30 - 17:00
(UTC +8)
04-09-20 Unsupervised Question Answering by Cloze Translation Paper,
Template-Based Question Generation from Retrieved Sentences for Improved Unsupervised Question Answering Paper
Slide,
Slide
Wang Weishi,
Wang Weishi
15:30 - 17:00
(UTC +8)
11-09-20 SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient Paper Slide Han Cheol 15:30 - 17:00
(UTC +8)
18-09-20 A Study of Reinforcement Learning for Neural Machine Translation Paper Slide Lin Xiang 15:30 - 17:00
(UTC +8)
07-10-20 An Actor-Critic Algorithm for Sequence Prediction Paper Slide Ankit 15:30 - 17:00
(UTC +8)
14-10-20 MLMLM: Link Prediction with Mean Likelihood Masked Language Model Paper Slide Ankit 15:30 - 17:00
(UTC +8)
21-10-20 Re-evaluating Evaluation in Text Summarization Paper Slide Lin Xiang 15:30 - 17:00
(UTC +8)
28-10-20 Self-training Improves Pre-training for Natural Language Understanding Paper Slide Tasnim 15:30 - 17:00
(UTC +8)
04-11-20 Improving Zero and Few-Shot Abstractive Summarization with Intermediate Fine-tuning and Data Augmentation Paper Slide Han Simeng 15:30 - 17:00
(UTC +8)

2020 Jan-May | Transfer Learning
Time: Thursdays 15:00 - 17:00

Date Topic Content Presenter(s) Time
23-01-20 kNN Language Model Paper Slide Shafiq 15:00 - 17:00
(UTC +8)

2019 Jun-Dec | Deep Learning for Structured Objects and Variational Methods
Time: Fridays 15:30 - 17:30

Date Topic Content Presenter(s) Time
21-06-19 GPT Paper,
XLM Paper,
Sparse Transformer Paper
Slide,
Slide,
Slide
Tasnim,
Tasnim,
Shafiq
15:30-17:30
(UTC +8)
28-06-19 Constituicy Parsing Paper,
Sampling Paper
Slide Slide Lin,
LinLin
15:30-17:30
(UTC +8)
05-07-19 Transformer-XL Slide Saiful 15:30-17:30
(UTC +8)
12-07-19 XL-Net Slide Phi 15:30-17:30
(UTC +8)
20-12-19 Mixture models for diverse MT Paper,
NTG with unlikelihood training Paper
Slide,
Slide
Phi,
Han
15:30-17:30
(UTC +8)

2019 Jan-May | Deep Learning for Structured Objects and Variational Methods
Time: Tuesdays 11:30 - 12:30

Date Topic Content Presenter(s) Time
22-01-19 Tree LSTM as an encoder Paper,
Doubly Recurrent net as a decoder Paper
Slide Shafiq 11:30-12:30
(UTC +8)
29-01-19 Top-down Tree structured Decoding
Paper
Slide Shafiq 11:30-12:30
(UTC +8)
12-02-19 Auto-Encoding Scene Graphs for Image Captioning
Paper
Slide Shafiq 11:30-12:30
(UTC +8)
05-04-19 Ordered Neuron Paper1,
Pay Less Attention Paper2
Slide1,
Slide2
Shafiq, Thomas,
Phi
15:30-17:30
(UTC +8)
12-04-19 BERT
Paper
Slide Shafiq 15:30-17:30
(UTC +8)
TBA Parsing with Recursive NNs
Paper
TBA Recursive Deep Models
Paper
TBA Constituency Parsing with a Self-Attentive Encoder
Paper

2018 March-November | Multilingual and Unsupervised Deep Learning

Date Topic Content Presenter(s) Time
07-03-2018 Recurrent Neural Network Part 1 Part 2 Tasnim, and Saiful 14:00-15:00
(UTC +8)
14-03-2018 Sequence to Sequence Learning
Paper 1, Paper 2
Part 1 Part 2 Tasnim, and Saiful 10:30-11:30
(UTC +8)
21-03-2018 Attention Mechanism
Paper 1, Paper 2
Part 1 Part 2 Tasnim, and Saiful 11:30-12:30
(UTC +8)
04-04-2018 Discourse Phenomena in NMT
Paper
Slide Phi 11:30-12:30
(UTC +8)
11-04-2018 Unsupervised Machine Translation Using Monolingual Corpora Only
Paper
Slide Song 11:30-12:30
(UTC +8)
25-04-2018 Fast and Accurate Entity Recognition with Iterated Dilated Convolutions
Paper
Slide Saiful 11:30-12:30
(UTC +8)
09-05-2018 Adversarial Domain Adaptation
Paper1, Paper2
Slide Shafiq 11:30-12:30
(UTC +8)
13-06-2018 Semi-Supervised Learning with Graph Embedding
Paper
Slide Tasnim 11:30-12:30
(UTC +8)
27-06-2018 Graph Convolutional Networks
Paper
Slide Song 11:30-12:30
(UTC +8)
11-07-2018 Transformer Network
Paper
Slide1, Slide2 Phi, and Shafiq 11:30-12:30
(UTC +8)
11-07-2018 Limitations of Unsup. Word Translation
Paper
Slide Song 11:30-12:30
(UTC +8)
13-09-2018 X-lingual NER
Paper
Slide Shafiq 12:00-13:00
(UTC +8)
04-10-2018 Context-Aware Pronoun Translation, and Word Sense Disambiguation Test Suite
Paper1, Paper2
Slide Jwala 12:00-13:00
(UTC +8)
11-10-2018 NMT of Rare Words with Subword Units
Paper
Slide Shafiq 12:00-13:00
(UTC +8)
18-10-2018 Pointer Networks
Paper
Slide Shafiq 10:00-11:00
(UTC +8)
08-11-2018 Tree Structured Decoding for NMT
Paper
12:00-13:00
(UTC +8)
TBA Neural MT and Image Caption Generation
Paper 1, Paper 2
TBA TBA
TBA Abstractive Summarization
Paper
TBA TBA