| Week | Date | Topic | Presenters | Papers | Slides |
|---|---|---|---|---|---|
| Week 3 | 1/21 | Biases in Language Processing | Shruti Sharan Anurag Pant |
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings Semantics derived automatically from language corpora contain human-like biases |
Link |
| Week 3 | 1/21 | Biases in Language Processing | Avijit Verma |
Understanding the Origins of Bias in Word Embeddings |
Link |
| Week 3 | 1/23 | Biases in Language Processing | Sepideh Parhami Doruk Karınca |
Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints Women Also Snowboard: Overcoming Bias in Captioning Models |
Link |
| Week 4 | 1/28 | Biases in Language Processing | Aakash Srinivasan Arvind krishna |
Bias in bios: A case study of semantic representation bias in a high-stakes setting What's in a Name? Reducing Bias in Bios without Access to Protected Attributes |
Link |
| Week 4 | 1/28 | Algorithmic Fairness | Paul Lou Pratyush Garg |
Equality of Opportunity in Supervised Learning |
Link |
| Week 4 | 1/30 | Algorithmic Fairness | Evan Czyzycki Anaelia Ovalle |
Fairness Through Awareness |
Link |
| Week 4 | 1/30 | Robustness in NLP | Vivek Krishnamurthy Daniel Ciao |
Adversarial Examples for Evaluating Reading Comprehension Systems. Semantically Equivalent Adversarial Rules for Debugging NLP models. |
Link |
| Week 5 | 2/4 | Robustness in NLP | Ritam Sarmah Sidharth Dhawan |
Universal Adversarial Triggers for Attacking and Analyzing NLP. Investigating Robustness and Interpretability of Link Prediction via Adversarial Modifications. |
Link |
| Week 5 | 2/4 | Robustness in NLP | Chong Zhang Simeng Pang |
Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation Certified Robustness to Adversarial Word Substitutions. |
Link |
| Week 6 | 2/11 | Inclusive NLP | Yizhao Wang Zhe Zhang |
Demographic Dialectal Variation in Social Media: A Case Study of African-American English Overcoming Language Variation in Sentiment Analysis with Social Attention |
Link |
| Week 6 | 2/13 | Data Bias and Domain Adaptation | Benlin Liu Xiaojian Ma |
Frustratingly Easy Domain Adaptation Strong Baselines for Neural Semi-supervised Learning under Domain Shift |
Link |
| Week 7 | 2/18 | Data Bias and Domain Adaptation | Yu-Chen Lin Jo-Chi Chuang |
Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasets The effect of different writing tasks on linguistic style: A case study of the ROC story cloze task |
Link |
| Week 7 | 2/18 | Data Bias and Domain Adaptation | Shiqi Wang Danfeng Guo |
Split and rephrase: Better evaluation and stronger baselines Performance impact caused by hidden bias of training data for recognizing textual entailment |
Link |
| Week 7 | 2/20 | Statistical significant test in NLP | Yuqing Wang |
The hitchhikers guide to testing statistical significance in natural language processing. An empirical investigation of statistical significance in nlp |
Link |
| Week 7 | 2/20 | Probing NLP Models | Arjun Subramonian John Arthur Dang |
Designing and Interpreting Probes with Control Tasks |
Link |
| Week 6 | 2/25 | Probing NLP Models | Xiangning Chen Yuanhao Xiong |
Do Neural NLP Models Know Numbers? Probing Numeracy in Embeddings Linguistic knowledge and transferability of contextual representations |
Link |
| Week 8 | 2/25 | Probing NLP Models | Shanxiu He Liunian Harold Li |
What you can cram into a single $&!#* vector: Probing sentence embeddings for linguistic properties A structural probe for finding syntax in word representations. |
Link |
| Week 8 | 2/27 | Probing NLP Models | Qingyi Zhao Spenser Wong |
What do neural machine translation models learn about morphology? |
Link |
| Week 8 | 2/27 | Probing NLP Models | Yining Hong Zhufeng Pan |
A causal framework for explaining the predictions of black-box sequence-to-sequence models |
Link |
| Week 9 | 3/3 | Explainable NLP | Yu Yang Roshni Iyer |
Attention is Not Not Explanation Attention is not explanation |
Link |
| Week 9 | 3/3 | Explain Data | Xuanqing Liu Difan Zou |
Data Shapley: Equitable Valuation of Data for Machine Learning |
Link |
| Week 9 | 3/5 | Machine Commonsense | Masoud Monajatipoor Li Cheng Lan |
ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning COMET: Commonsense Transformers for Knowledge Graph Construction |
Link |
| Week 9 | 3/5 | Machine Commonsense | Mukund Mundhra |
Event2Mind: Commonsense Inference on Events, Intents, and Reactions SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference |
Link |
| Week 10 | 3/10 | Final Project Presentaiton | |||
| Week 12 | 3/10 | Final Project Presentaiton |
| Name | Papers | Link |
|---|---|---|
| Paul Lou | Inherent Trade-Offs in the Fair Determination of Risk Scores | Link |
| Shiqi Wang | Discriminative Training Methods for Hidden Markov Models with Perceptron Algorithms: An Introduction | Link |
| Xiangning Chen | The Evolved Transformer | Link |
| Chong Zhang | Query-Efficient Hard-label Black-box Attack: An Optimization-based Approach | Link |
| Yizhao Wang | A Single-Channel Noise Reduction Filtering/Smoothing Technique In The Time Domain | Link |
| Ritam Sarmah | Declarative Question Answering over Knowledge Bases Containing Natural Language Text with Answer Set Programming | Link |
| Yuanhao Xiong | Man is to Computer Programmer as Woman is to Homemaker? Debiasing word embedding: An Introduction to Biases in Natural Language Processing | Link |
| Anaelia Ovalle | Mitigating Gender Bias in Natural Language Processing: Literature Review | Link |
| Evan Czyzycki | A Benchmark for Interpretability Methods in Deep Neural Networks | Link |
| Yu-Chen Lin | Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints | Link |
| Simeng Pang | Joint Embeddings of Chinese Words, Characters and Fine-grained Subcharacter Components | Link |
| Yuqing Wang | Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment | Link |
| Qingyi Zhao | Detecting Political Bias in News Articles Using Headline Attention | Link |
| Jo-Chi Chuang | Unsupervised cross-modal alignment of speech and text embedding spaces | Link |
| Danfeng Guo | Attention is not not Explanation | Link |
| Arjun Subramonian | Learning Neural Templates for Text Generation | Link |
| Difan Zou | Certified robustness to adversarial word substitutions. | Link |
| Roshni Iyer | Transformer-XH: Multi-hop question answering with eXtra Hop Attention | Link |
| Yining Hong | Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision | Link |
| Shruti Sharan | Generating Natural Language Adversarial Examples | Link |
| Xuanqing Liu | FreeLB: Enhanced Adversarial Training for Language Understanding | Link |
| Aakash Srinivasan | Lipstick on a Pig — Existing Debiasing Methods Simply Cover-up Systematic Gender Biases in Word Embeddings But do not Remove Them | Link |
| Spenser Wong | Attention is All You Need | Link |
| Daniel Ciao | Visualizing and Measuring the Geometry of BERT | Link |
| Vivek Krishnamurthy | Attention is All You Need | Link |
| Anurag Pant | Explaining Fine-Grained Detection of Propaganda in News Articles using NLP | Link |
| Xiaojian Ma | Grad-CAM: Visual Explanations from Deep Networksvia Gradient-based Localization | Link |
| Benlin Liu | On the Robustness of Self-Attentive Models | Link |
| Masoud Monajatipoor | few-Shot Representation Learning for Out-Of-Vocabulary Words | Link |
| Arvind krishna Sridhar | Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints | Link |
| Li Cheng Lan | Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks | Link |
| Pratyush Garg | Balancing Fairness and Accuracy in Decision Making | Link |
| Yu Yang | Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations | Link |
| Sidharth Dhawan | Attention is All You Need | Link |
| Doruk Karinca | Don’t Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases | Link |
| Zhe Zhang | Emotion-Cause Pair Extraction: A New Task to Emotion Analysis in Texts | Link |
| Zhufeng Pan | Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer | Link |
| Mukund Mundhra | Augmenting End-to-End Dialogue Systems with Commonsense Knowledge | Link |
| Liunian Harold Li | Deep Compositional Question Answering with Neural Module Networks | Link |
| Sepideh Parhami | Angry stomach is angry - or how we can translate tweet-speak for the medical world | Link |
| Avijit Verma | Automatically Neutralizing Subjective Bias in Text | Link |
| Shanxiu He | BERT Rediscovers the Classical NLP Pipeline | Link |
| John Arthur Dang | Attention is All You Need | Link |