Dejiao Zhang

About Me

I’m a senior applied scientist at AWS AI Labs, working on large language models (LLM) and their applications. I am part of the science lead team that builds Amazon CodeWhisperer. Before joining AWS, I received my Ph.D. degree in the Electrical Engineering and Computer Science department from the University of Michigan in 2019, under the supervision of Prof. Laura Balzano.

Selected Publications

• Reasoning in Token Economies: Budget-Aware Evaluation of LLM Reasoning Strategies

  Conference on Empirical Methods in Natural Language Processing, EMNLP 2024.

  Junlin Wang, Siddhartha Jain, Dejiao Zhang, Ben Athiwaratkun, et.al

• Repoformer: Selective Retrieval for Repository-level Code Completion

  International Conference on Machine Learning, ICML 2024. (Oral, 1.5%) | Code | BlogPost

  Di Wu, Wasi Uddin Ahmad, Dejiao Zhang, et.al

• CodeSage: Code Representation Learning At Scale

  International Conference on Learning Representations, ICLR 2024. | HuggingFace | Code | BlogPost

  Dejiao Zhang*, Wasi Uddin Ahmad*, Ming Tan, et.al

• ContraCLM: Contrastive Learning for Causal Language Model

  Association for Computational Linguistics: ACL 2023 | Code

  Nihal Jain*, Dejiao Zhang*, Wasi Uddin Ahmad*, et.al

• Multitask Pretraining with Structured Knowledge for Text-to-SQL Generation

  Association for Computational Linguistics: ACL 2023

  Robert Giaquinto, Dejiao Zhang, Benjamin Kleiner, et.al

• Learning Dialogue Representations from Consecutive Utterances

  North American Chapter of the Association for Computational Linguistics, NAACL 2022. | HuggingFace | Code

  Zhihan Zhou, Dejiao Zhang, Wei Xiao, et.al

• Lifelong Pretraining: Continually Adapting Language Models to Emerging Corpora

  North American Chapter of the Association for Computational Linguistics, NAACL 2022. (Oral)

  Xisen Jin, Dejiao Zhang, Henghui Zhu, et.al

• Virtual Augmentation Supported Contrastive Learning of Sentence Representations

  Findings of the Association for Computational Linguistics: ACL 2022. | HuggingFace | Code

  Dejiao Zhang, Wei Xiao, Henghui Zhu, et.al

• Pairwise Supervised Contrastive Learning of Sentence Representations

  Conference on Empirical Methods in Natural Language Processing, EMNLP 2021. | HuggingFace | Code

  Dejiao Zhang, Sheng-Wen Li, Wei Xiao, et.al

• Supporting Clustering with Contrastive Learning

  North American Chapter of the Association for Computational Linguistics, NAACL 2021. | Code

  Dejiao Zhang, Feng Nan, Xiaokai Wei, et.al

• Margin-aware Unsupervised Domain Adaptation for Cross-lingual Text Labeling

  Findings of the Association for Computational Linguistics: EMNLP 2020.

  Dejiao Zhang, Ramesh Nallapati, Henghui Zhu, et.al

• Learning to Share: Simultaneous Parameter Tying and Sparsification in Deep Learning

  International Conference on Learning Representations, ICLR 2018. | Code

  Dejiao Zhang*, Haozhu Wang*, Mario A.T. Figueiredo, Laura Balzano.

• Global Convergence of a Grassmannian Gradient Descent Algorithm for Subspace Estimation

  Artificial Intelligence and Statistics, AISTATS 2016.

  Dejiao Zhang, Laura Balzano.

• Iterative Grassmannian Optimization for Robust Image Alignment

  Image and Vision Computing, 2014.

  Jun He, Dejiao Zhang, Laura Balzano, Tao Tao.