About Me
Jun Li is an Associate Professor in the Department of Computer Science at Queens College and a doctoral faculty in the Graduate Center, City University of New York (CUNY). Before joining CUNY, he was an Assistant Professor at the School of Computing and Information Sciences, Florida International University. He received his Ph.D. in Electrical and Computer Engineering from the University of Toronto in 2017, and his B.S. and M.S. degrees in Computer Science from Fudan University, China, in 2009 and 2012, respectively.
Merging the gap between theory and practice, his research explores how large language models can be systematically evaluated, strategically applied, and information-theoretically understood — spanning LLM evaluation methodology, semantic text compression, LLM-powered input methods, and AI-augmented computer science education. Building on his earlier work in coding theory for distributed systems and large-scale machine learning, his research bridges formal information theory with modern language AI. His work has been supported by the National Science Foundation, Google, AWS, and CUNY.
Research Interests
Large language models: evaluation, reasoning, and applications in education
Semantic information theory and LLM-based text compression
LLM-powered input methods and multilingual NLP
Coding theory and erasure codes for distributed systems
Distributed storage systems and data center architectures
Publications
Author names in italics are students in my group.
2026
- arXiv 2026Taxonomy-Targeted Error Generation for Quantitative ReasoningarXiv:2605.29007, May 2026.
- ITW 2026Pipelined Gradient CodingIEEE Information Theory Workshop (ITW), November 2026.
- AACL-IJCNLP 2026Text-Preserving Lossy Text Compression: A Study of Strategic Deletion and LLM ReconstructionInternational Joint Conference on Natural Language Processing & Asia-Pacific Chapter of the Association for Computational Linguistics (AACL-IJCNLP), Main Conference, Hengqin, China, November 2026.
- IWQoS 2026FedCod: An Efficient Coded Communication Protocol for Cross-Silo Federated LearningIEEE/ACM International Symposium on Quality of Service (IWQoS), June 2026.
2025
2024
- ICC 2024Sequence-aware Coding for Matrix Multiplication with Arbitrary RecoverabilityIEEE International Conference on Communications (ICC), June 2024.
2023
- SC 2023Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data CentersInternational Conference for High Performance Computing, Networking, Storage, and Analysis (SC), November 2023.
- ICDCS 2023On Arbitrary Ignorance of Stragglers with Gradient Coding43rd IEEE International Conference on Distributed Computing Systems (ICDCS), July 2023.
2022
2021
- IEEE TCC 2021Demand-aware Erasure Coding for Distributed Storage SystemsIEEE Transactions on Cloud Computing, vol. 9, no. 2, pp. 532-545, 2021.
- IWQoS 2021Coded Matrix Chain Multiplication2021 IEEE/ACM 28th International Symposium on Quality of Service (IWQoS), June 2021.
2020
- ISIT 2020Straggler-free Coding for Concurrent Matrix Multiplications2020 IEEE International Symposium on Information Theory (ISIT), June 2020.
- ISIT 2020Local Re-encoding for Coded Matrix Multiplication2020 IEEE International Symposium on Information Theory (ISIT), June 2020.
2019
- IEEE TETC 2019Mist: Efficient Dissemination of Erasure-coded Data in Data CentersIEEE Transactions on Emerging Topics in Computing, vol. 7, no. 3, pp. 468-480, 2019.
2018
- ICDCS 2018Parallelism-Aware Locally Repairable Code for Distributed Storage SystemsIEEE International Conference on Distributed Computing Systems (ICDCS), July 2018.
2017
- IEEE TPDS 2017Beehive: Erasure Codes for Fixing Multiple Failures in Distributed Storage SystemsIEEE Transactions on Parallel and Distributed Systems, vol. 28, no. 5, pp. 1257-1270, May 2017.
- ICDCS 2017On Data Parallelism of Erasure Coding in Distributed Storage SystemsIEEE International Conference on Distributed Computing Systems (ICDCS), June 2017.
2016
- SC 2016Multi-Resource Fair Sharing for Datacenter Jobs with Placement ConstraintsThe International Conference for High Performance Computing, Networking, Storage and Analysis (SC), November 2016.
- IWQoS 2016Zebra: Demand-aware Erasure Coding for Distributed Storage SystemsIEEE/ACM International Symposium on Quality of Service (IWQoS), June 2016.
- SIGMETRICS 2016Towards Multi-Resource Fair Allocation with Placement ConstraintsACM SIGMETRICS 2016 (2-page poster paper), June 2016.
2015
- HotStorage 2015Beehive: Erasure Codes for Fixing Multiple Failures in Distributed Storage SystemsUSENIX Workshop on Hot Topics in Storage and File Systems (HotStorage), July 2015.
2014
- INFOCOM 2014Cooperative Repair with Minimum-Storage Regenerating Codes for Distributed StorageIEEE Conference on Computer Communications (INFOCOM), April 2014.
2013
- TST 2013Erasure Coding for Cloud Storage Systems: A SurveyTsinghua Science and Technology, vol. 18, no. 3, pp. 259-272, June 2013.
- INFOCOM 2013Cooperative Pipelined Regeneration in Distributed Storage SystemsIEEE Conference on Computer Communications (INFOCOM), April 2013.
2011
2010
2009
- MSR TR 2009The Generalized DCell Network Structures and Their Graph PropertiesMicrosoft Research TechReport, MSR-TR-2009-140.