🏢 Works of Microsoft from 2021

16 works · the most cited first · ← Microsoft

1. DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding SharingarXiv:2111.09543Pengcheng He, Jianfeng Gao, Weizhu Chen2021 · 1,316 citations
2. Are NLP Models really able to Solve Simple Math Word Problems?Arkil Patel, Satwik Bhattamishra, Navin Goyal2021 · 883 citations
3. LayoutLMv2: Multi-modal Pre-training for Visually-rich Document UnderstandingAnnual Meeting of the Association for Computational Linguistics (ACL)Yang Xu, Yiheng Xu, Tengchao Lv, et al.2021 · 623 citations
4. QMSum: A New Benchmark for Query-based Multi-domain Meeting SummarizationConference of the North American Chapter of the Association for Computational Linguistics (NAACL)Ming Zhong, Da Yin, Tao Yu, et al.2021 · 400 citations
5. GraphCodeBERT: Pre-training Code Representations with Data FlowarXiv:2009.08366Daya Guo, Shuo Ren, Shuai Lu, et al.2021 · 331 citations
6. ZeRO-Offload: Democratizing Billion-Scale Model TrainingarXiv:2101.06840Jie Ren, Samyam Rajbhandari, Reza Yazdani Aminabadi, et al.2021 · 303 citations
7. Want To Reduce Labeling Cost? GPT-3 Can HelpFindings of the Association for Computational LinguisticsShuohang Wang, Yang Liu, Yichong Xu, et al.2021 · 298 citations
8. Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving ModelsAnnual Meeting of the Association for Computational Linguistics (ACL)Tongshuang Wu, Marco Tulio Ribeiro, Jeffrey Heer, et al.2021 · 272 citations
9. TAPEX: Table Pre-training via Learning a Neural SQL ExecutorarXiv:2107.07653Qian Liu, Bei Chen, Jiaqi Guo, et al.2021 · 254 citations
10. The Curious Case of Hallucinations in Neural Machine TranslationConference of the North American Chapter of the Association for Computational Linguistics (NAACL)Vikas Raunak, Arul Menezes, Marcin Junczys-Dowmunt2021 · 241 citations
11. Challenges in Detoxifying Language ModelsFindings of the Association for Computational LinguisticsJohannes Welbl, Amelia Glaese, Jonathan Uesato, et al.2021 · 207 citations
12. COCO-LM: Correcting and Contrasting Text Sequences for Language Model PretrainingarXiv:2102.08473Yu Meng, Chenyan Xiong, Payal Bajaj, et al.2021 · 196 citations
13. Enhancing Factual Consistency of Abstractive SummarizationConference of the North American Chapter of the Association for Computational Linguistics (NAACL)Chenguang Zhu, William Hinthorn, Ruochen Xu, et al.2021 · 188 citations
14. MediaSum: A Large-scale Media Interview Dataset for Dialogue SummarizationConference of the North American Chapter of the Association for Computational Linguistics (NAACL)Chenguang Zhu, Yang Liu, Jie Mei, et al.2021 · 180 citations
15. Structure-Grounded Pretraining for Text-to-SQLConference of the North American Chapter of the Association for Computational Linguistics (NAACL)Xiang Deng, Ahmed Hassan Awadallah, Christopher Meek, et al.2021 · 166 citations
16. Joint Universal Syntactic and Semantic ParsingTransactions of the Association for Computational Linguistics (TACL)Elias Stengel-Eskin, Kenton Murray, Sheng Zhang, et al.2021 · 15 citations

This page in Markdown → https://md.research.elamur.ai/free/en/companies/microsoft/year/2021/