The Conference on Empirical Methods in Natural Language Processing (EMNLP) 2020 is being hosted virtually from November 16th – November 20th. We’re excited to share all the work from SAIL that’s being presented, and you’ll find links to papers, videos and blogs below. Feel free to reach out to the contact authors directly to learn more about the work that’s happening at Stanford!
Main Conference
Pre-Training Transformers as Energy-Based Cloze Models
Authors: Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning
Contact: kevclark@cs.stanford.edu
Links: Paper
Keywords: representation learning, self-supervised learning, energy-based models
ALICE: Active Learning with Contrastive Natural Language Explanations
Authors: Weixin Liang, James Zou, Zhou Yu
Contact: wxliang@stanford.edu
Links: Paper
Keywords: natural language explanation, class-based active learning, contrastive explanation
CheXbert: Combining Automatic Labelers and Expert Annotations for Accurate Radiology Report Labeling Using BERT
Authors: Akshay Smit, Saahil Jain, Pranav Rajpurkar, Anuj Pareek, Andrew Y. Ng, Matthew P. Lungren
Contact: akshaysm@stanford.edu
Links: Paper
Keywords: bert, natural language processing, radiology, medical imaging, deep learning
AutoQA: From Databases To QA Semantic Parsers With Only Synthetic Training Data
Authors: Silei Xu, Sina J. Semnani, Giovanni Campagna, Monica S. Lam
Contact: silei@cs.stanford.edu
Links: Paper
Keywords: question answering, semantic parsing, language models, synthetic training data, data augmentation
Data and Representation for Turkish Natural Language Inference
Authors: Emrah Budur, Rıza Özçelik, Tunga Güngör, Christopher Potts
Contact: emrah.budur@boun.edu.tr
Links: Paper | Website
Keywords: sentence-level semantics, natural language inference, neural machine translation, morphologically rich language
Intrinsic Evaluation of Summarization Datasets
Authors: Rishi Bommasani, Claire Cardie
Contact: nlprishi@stanford.edu
Links: Paper | Video | Website | Virtual Conference Room
Keywords: summarization, datasets, evaluation
Learning Music Helps You Read: Using Transfer to Study Linguistic Structure in Language Models
Authors: Isabel Papadimitriou, Dan Jurafsky
Contact: isabelvp@stanford.edu
Links: Paper
Keywords: transfer learning, analysis, music, hierarchical structure
Localizing Open-Ontology QA Semantic Parsers in a Day Using Machine Translation
Authors: Mehrad Moradshahi, Giovanni Campagna, Sina J. Semnani, Silei Xu, Monica S. Lam
Contact: mehrad@cs.stanford.edu
Links: Paper | Website
Keywords: machine translation, semantic parsing, localization
SLM: Learning a Discourse Language Representation with Sentence Unshuffling
Authors: Haejun Lee, Drew A. Hudson, Kangwook Lee, Christopher D. Manning
Contact: dorarad@stanford.edu
Links: Paper
Keywords: transformer, bert, language, understanding, nlp, squad, glue, sentences, discourse
Utility is in the Eye of the User: A Critique of NLP Leaderboards
Authors: Kawin Ethayarajh, Dan Jurafsky
Contact: kawin@stanford.edu
Links: Paper | Website
Keywords: nlp, leaderboard, utility, benchmark, fairness, efficiency
With Little Power Comes Great Responsibility
Authors: Dallas Card, Peter Henderson, Urvashi Khandelwal, Robin Jia, Kyle Mahowald, Dan Jurafsky
Contact: dcard@stanford.edu
Links: Paper | Website
Keywords: statistical power, experimental methodology, leaderboards, machine translation, human evaluation
Findings of EMNLP
DeSMOG: Detecting Stance in Media On Global Warming
Authors: Yiwei Luo, Dallas Card, Dan Jurafsky
Contact: yiweil@stanford.edu
Links: Paper | Website
Keywords: computational social science; framing; argumentation; stance; bias; climate change
Investigating Transferability in Pretrained Language Models
Authors: Alex Tamkin, Trisha Singh, Davide Giovanardi, Noah Goodman
Contact: atamkin@stanford.edu
Links: Paper | Website | Virtual Conference Room
Keywords: finetuning, transfer learning, language models, bert, probing
Stay Hungry, Stay Focused: Generating Informative and Specific Questions in Information-Seeking Conversations
Authors: Peng Qi, Yuhao Zhang, Christopher D. Manning
Contact: pengqi@cs.stanford.edu
Links: Paper | Blog Post
Keywords: conversational agents, question generation, natural language generation
Do Language Embeddings Capture Scales?
Authors: Xikun Zhang*, Deepak Ramachandran*, Ian Tenney, Yanai Elazar, Dan Roth
Contact: xikunz2@cs.stanford.edu
Links: Paper | Virtual Conference Room
Keywords: probing, analysis, bertology, scales, common sense knowledge
On the Importance of Adaptive Data Collection for Extremely Imbalanced Pairwise Tasks
Authors: Stephen Mussmann, Robin Jia, Percy Liang
Contact: robinjia@cs.stanford.edu
Links: Paper | Website
Keywords: active learning, robustness, label imbalance
Pragmatic Issue-Sensitive Image Captioning
Authors: Allen Nie, Reuben Cohn-Gordon, Christopher Potts
Contact: anie@stanford.edu
Links: Paper | Video
Keywords: controllable caption generation, question under discussion, discourse, pragmatics
Workshops and Co-Located Conferences
BLEU Neighbors: A Reference-less Approach to Automatic Evaluation
Authors: Kawin Ethayarajh, Dorsa Sadigh
Contact: kawin@stanford.edu
Links: Paper | Website
Keywords: nlp, bleu, evaluation, nearest neighbors, dialogue
Determining Question-Answer Plausibility in Crowdsourced Datasets Using Multi-Task Learning
Authors: Rachel Gardner, Maya Varma, Clare Zhu, Ranjay Krishna
Contact: rachel0@cs.stanford.edu
Links: Paper
Keywords: noisy text, bert, plausibility, multi-task learning
Explaining the ‘Trump Gap’ in Social Distancing Using COVID Discourse
Authors: Austin van Loon, Sheridan Stewart, Brandon Waldon, Shrinidhi K. Lakshmikanth, Ishan Shah, Sharath Chandra Guntuku, Garrick Sherman, James Zou, Johannes Eichstaedt
Contact: avanloon@stanford.edu
Links: Paper
Keywords: computational social science, social distancing, word2vec, vector semantics, twitter, bert
Learning Adaptive Language Interfaces through Decomposition
Authors: Siddharth Karamcheti, Dorsa Sadigh, Percy Liang
Contact: skaramcheti@cs.stanford.edu
Links: Paper | Virtual Conference Room
Keywords: semantic parsing, interaction, decomposition
Modeling Subjective Assessments of Guilt in Newspaper Crime Narratives
Authors: Elisa Kreiss*, Zijian Wang*, Christopher Potts
Contact: ekreiss@stanford.edu
Links: Paper | Website
Keywords: psycholinguistics, pragmatics, token-level supervision, model attribution, news, guilt, hedges, corpus, subjectivity
Neural Natural Language Inference Models Partially Embed Theories of Lexical Entailment and Negation
Authors: Atticus Geiger, Kyle Richardson, Chris Potts
Contact: atticusg@stanford.edu
Links: Paper | Website
Keywords: entailment intervention causality systematic generalization
Structured Self-Attention Weights Encode Semantics in Sentiment Analysis
Authors: Zhengxuan Wu, Thanh-Son Nguyen, Desmond C. Ong
Contact: wuzhengx@stanford.edu
Links: Paper
Keywords: attention, explainability, sentiment analysis
We look forward to seeing you at EMNLP 2020!