{"659850":{"#nid":"659850","#data":{"type":"news","title":"Latest NLP Research Derives Insight from Growing Volume of Digital Text","body":[{"value":"\u003Cp\u003ENew NLP research from Georgia Tech is allowing for patterns to be uncovered in this text and broaden the understanding of how to build better computer applications that derive value from written language.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EGeorgia Tech researchers are presenting their latest work at the annual conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022), taking place this week, July 10-15. NAACL provides a regional focus for members of the Association for Computational Linguistics (ACL) in North America as well as in Central and South America and promotes cooperation and information exchange among related scientific and professional societies.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;Recent advances in natural language processing \u0026shy;\u0026shy;\u0026shy;\u0026shy;\u0026ndash; especially around big models \u0026ndash; have enabled successful applications,\u0026rdquo; said\u0026nbsp;\u003Cstrong\u003EDiyi Yang\u003C\/strong\u003E, assistant professor in the School of Interactive Computing and researcher in NLP. \u0026ldquo;At the same time, we see a growing amount of evidence and concern toward the negative aspects of NLP systems, such as the bias and fragility exhibited by these models, as well as the lack of input from users.\u0026rdquo;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EYang\u0026rsquo;s work in computational social science and NLP focuses on how to understand human communication in social context\u0026nbsp;and build\u0026nbsp;socially aware\u0026nbsp;language technologies to support human-to-human and human-computer interaction.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EHer SALT Lab has accrued an impressive number of innovations in the field over the past eight months, starting with research presented at last November\u0026rsquo;s EMNLP conference. SALTers, as they are called, led Georgia Tech to become the top global contributor in computational social science and cultural analytics at that venue. The 60th\u0026nbsp;Meeting of the ACL in Dublin followed in May with multiple SALT studies, including a best paper. Yang\u0026rsquo;s group has six papers at this week\u0026rsquo;s NAACL.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;We hope to build NLP systems that are more user centric, more robust, and more aware of human factors,\u0026rdquo; said Yang. \u0026ldquo;Our NAACL works are in this direction, covering robustness, toxicity detection, and generalization to new settings.\u0026rdquo;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EYang\u0026rsquo;s aspirations for the field are shared by her Tech peers, who have work in the following tracks at NAACL:\u003C\/p\u003E\r\n\r\n\u003Cul\u003E\r\n\t\u003Cli\u003EEthics, Bias, Fairness\u003C\/li\u003E\r\n\t\u003Cli\u003EInformation Extraction\u003C\/li\u003E\r\n\t\u003Cli\u003EInformation Retrieval\u003C\/li\u003E\r\n\t\u003Cli\u003EInterpretability and Analysis of Models for NLP\u003C\/li\u003E\r\n\t\u003Cli\u003EMachine Learning\u003C\/li\u003E\r\n\t\u003Cli\u003EMachine Learning for NLP\u003C\/li\u003E\r\n\t\u003Cli\u003ESemantics: Sentence-level Semantics and Textual Inference\u003C\/li\u003E\r\n\u003C\/ul\u003E\r\n\r\n\u003Cp\u003EGeorgia Tech\u0026rsquo;s research paper acceptances in the main program at NAACL are below. To learn more about NLP and machine learning research at Georgia Tech visit\u0026nbsp;\u003Ca href=\u0022https:\/\/ml.gatech.edu\/\u0022\u003Ehttps:\/\/ml.gatech.edu\u003C\/a\u003E.\u003C\/p\u003E\r\n\r\n\u003Ch4\u003E\u003Cstrong\u003EGEORGIA TECH RESEARCH AT NAACL 2022\u003C\/strong\u003E\u0026nbsp;(main papers program)\u003C\/h4\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EEthics, Bias, Fairness\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EExplaining Toxic Text via Knowledge Enhanced Text Generation\u003Cbr \/\u003E\r\n\u003Cem\u003ERohit Sridhar, Diyi Yang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EInformation Extraction\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESelf-Training with Differentiable Teacher\u003Cbr \/\u003E\r\n\u003Cem\u003ESimiao Zuo, Yue Yu, Chen Liang, Haoming Jiang, Siawpeng Er, Chao Zhang, Tuo Zhao, Hongyuan Zha\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EInformation Retrieval\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ECERES: Pretraining of Graph-Conditioned Transformer for Semi-Structured Session Data\u003Cbr \/\u003E\r\n\u003Cem\u003ERui Feng, Chen Luo, Qingyu Yin, Bing Yin, Tuo Zhao, Chao Zhang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EInterpretability and Analysis of Models for NLP\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIdentifying and Mitigating Spurious Correlations for Improving Robustness in NLP Models\u003Cbr \/\u003E\r\n\u003Cem\u003ETianlu Wang, Rohit Sridhar, Diyi Yang, Xuezhi Wang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EMeasure and Improve Robustness in NLP Models: A Survey\u003Cbr \/\u003E\r\n\u003Cem\u003EXuezhi Wang, Haohan Wang, Diyi Yang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EReframing Human-AI Collaboration for Generating Free-Text Explanations\u003Cbr \/\u003E\r\n\u003Cem\u003ESarah Wiegreffe, Jack Hessel, Swabha Swayamdipta, Mark Riedl, Yejin Choi\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EMachine Learning\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EAcTune: Uncertainty-Aware Active Self-Training for Active Fine-Tuning of Pretrained Language Models\u003Cbr \/\u003E\r\n\u003Cem\u003EYue Yu, Lingkai Kong, Jieyu Zhang, Rongzhi Zhang, Chao Zhang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EMoEBERT: from BERT to Mixture-of-Experts via Importance-Guided Adaptation\u003Cbr \/\u003E\r\n\u003Cem\u003ESimiao Zuo, Qingru Zhang, Chen Liang, Pengcheng He, Tuo Zhao, Weizhu Chen\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EMachine Learning for NLP\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ETreeMix: Compositional Constituency-based Data Augmentation for Natural Language Understanding\u003Cbr \/\u003E\r\n\u003Cem\u003ELe Zhang, Zichao Yang, Diyi Yang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ENLP Applications\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ECryptocoin Bubble Detection: A New Dataset, Task \u0026amp; Hyperbolic Models\u003Cbr \/\u003E\r\n\u003Cem\u003ERamit Sawhney, Shivam Agarwal, Vivek Mittal, Paolo Rosso, Vikram Nanda, Sudheer Chava\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ESemantics: Sentence-level Semantics and Textual Inference\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESEQZERO: Few-shot Compositional Semantic Parsing with Sequential Prompts and Zero-shot Models\u003Cbr \/\u003E\r\n\u003Cem\u003EJingfeng Yang, Haoming Jiang, Qingyu Yin, Danqing Zhang, Bing Yin, Diyi Yang\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESUBS: Subtree Substitution for Compositional Semantic Parsing\u003Cbr \/\u003E\r\n\u003Cem\u003EJingfeng Yang, Le Zhang, Diyi Yang\u003C\/em\u003E\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Natural language processing (NLP) is a growing cornerstone of artificial intelligence and allows people and machines to act based on insights gleaned from digital text."}],"uid":"35403","created_gmt":"2022-08-02 14:38:06","changed_gmt":"2022-08-02 14:38:06","author":"Carly Ralston","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2022-07-12T00:00:00-04:00","iso_date":"2022-07-12T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"659849":{"id":"659849","type":"image","title":"Diyi Yang","body":null,"created":"1659450976","gmt_created":"2022-08-02 14:36:16","changed":"1659450976","gmt_changed":"2022-08-02 14:36:16","alt":"","file":{"fid":"250096","name":"Diyi_Yang.jpeg","image_path":"\/sites\/default\/files\/images\/Diyi_Yang.jpeg","image_full_path":"http:\/\/tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/Diyi_Yang.jpeg","mime":"image\/jpeg","size":6955,"path_740":"http:\/\/tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/Diyi_Yang.jpeg?itok=QNu3lMh-"}}},"media_ids":["659849"],"groups":[{"id":"545781","name":"Institute for Data Engineering and Science"}],"categories":[],"keywords":[{"id":"187023","name":"go-data"}],"core_research_areas":[{"id":"39431","name":"Data Engineering and Science"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022https:\/\/www.cc.gatech.edu\/author\/joshua-preston\u0022\u003EJOSHUA PRESTON\u003C\/a\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}