Look Who’s Talking: Stanford Linguists at ACL-IJCNLP 2021

Many Stanford linguists presented their research at the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP) held on August 2-5, 2021:

Chris Potts gave a keynote talk titled “Reliable characterizations of NLP systems as a social responsibility”

Kawin Ethayarajh and Dan Jurafsky presented the project “Attention Flows are Shapley Value Explanations”

Dora Demszky, Dan Jurafsky, and Tatsunori Hashimoto, as well as Stanford GSE alums Jing Liu (Ph.D. ’18) and Julie Cohen (Ph.D. ’13) were among the presenters of the project “Measuring Conversational Uptake: A Case Study on Student-Teacher Interactions”

Chris Potts, Zhengxuan Wu, and Atticus Geiger were among the presenters of “DynaSent: A Dynamic Benchmark for Sentiment Analysis”

Omar Khattab, Chris Potts and Matei Zaharia presented “Relevance-guided Supervision for OpenQA with ColBERT”

Siddharth Karamcheti, Ranjay Krishna, Fei-Fei Li and Christopher Manning presented “Mind Your Outliers! Investigating the Negative Impact of Outliers on Active Learning for Visual Question Answering”