I am an assistant professor in Computer Science at North Carolina State University. I work on trustworthy, interpretable, and efficient AI/deep learning. My interest lies in the intersection of failure mode/safety risk/vulnerability/bias and efficiency of deep learning. I also have a background in safety-critical systems. Prior to joining NC State in 2022, I was an assistant professor in EECS at Syracuse University (2021-2022.) Before then, I was an associate research scientist in Computer Science at Yale University. I received my PhD in Computer Science at the University of Illinois at Urbana-Champaign, and my BS and MS degrees in Computer Science and Engineering at Seoul National University, Seoul, Korea.

 

jung-eun.kim@ncsu.edu

Available Positions

I am looking for genuinely motivated PhD students. If you are interested in working with me for a PhD program, please contact me with your CV and mention your research interests and experiences with regard to mine.

 

Research Focus of Our Group

In our group, we care about trustworthy, interpretable, and efficient AI/deep learning. We identify how models can go wrong and answer why. We fundamentally anatomize neural networks to understand and verify what is invariant in them and what causes failure modes/safety risks/vulnerabilities/biases. Deep learning is used to be generalized for high-dimensional problems – if you already know what to encounter, you must program it instead. However, such aforementioned issues hurt the generalizability and exacerbate memorization behaviors of AI/deep learning models. Also, when these issues meet efficiency considerations, we may gain or lose something. So, like a heart surgeon, we open the heart of a neural network architecture, look into it, interpret it, and cure it. This is our mission, and we are savvy about them.

 

Media Coverage

• New Technique Overcomes Spurious Correlations Problem in AI
• New Method Forecasts Computation, Energy Costs for Sustainable AI Models
  (Highlighted in the ACM Tech News)
• Take Aim: The Hottest Problems in Artificial Intelligence
• Researchers Show How Network Pruning can Skew Deep Learning Models

 

Selected Honors and Awards

• ICLR Spotlight, 2025
• IBM Faculty award, 2023
• CRA Early & Mid Career Mentoring Workshop, 2023
• Cloud GPU provided by Lambda, worth $17,280, for my course, Resource-dependent neural networks, Spring 2023. Thank you, Lambda!
• NeurIPS Spotlight and a nomination for Best Paper Award, 2022
• CRA (Computing Research Association) Career Mentoring Workshop, 2022
• NSF SaTC (Secure and Trustworthy Cyberspace): CORE: Small: Partition-Oblivious Real-Time Hierarchical Scheduling, Co-PI, National Science Foundation, 2020–2024
• GPU Grant by NVIDIA Corporation, 2018
• The MIT EECS Rising Stars, 2015
• The Richard T. Cheng Endowed Fellowship, 2015 – 2016

Selected Program Committee/Panel Service

• Program Committee/Reviewer of ICLR 2024-2025, ICML 2024-2025, NeurIPS 2023-2025, AAAI 2023-2025, IJCAI 2023-2025
• Publicity Chair of IJCAI 2024
• Senior Program Committee of AAAI 2024, Safe, Robust and Responsible AI track
• NSF review panel 2023, twice

 

Students

I am fortunate to advise and work with the brilliant students who have a vision for the future:

• Xingli Fang
• Varun Mulchandani
• Vishwesh Sangarya
• Jianwei Li
• Rishi Singhal
• Minseon Kim

Teaching

• Deep learning beyond accuracy, CSC 591 & 791 ECE 591, Fall 2023, Fall 2024, Spring 2025, Fall 2025
• Trustworthy and efficient deep learning, CSC 495 & 591, Spring 2025
• Resource-dependent neural networks, Spring 2023, Cloud GPU provided by Lambda, worth $17,280. Thank you, Lambda!
• Resource-/Time-dependent learning, Fall 2022

 

Publications

(* Students whom I advised are underlined.)

• “Safety Alignment Can be Not Superficial with Explicit Safety Signals”
  Jianwei Li and Jung-Eun Kim
  ICML 2025
  (This work is about LLM’s Safety Alignment.)
  [arXiv] [project website]
• “Severing Spurious Correlations with Data Pruning”
  Varun Mulchandani and Jung-Eun Kim
  ICLR 2025  🏆 Spotlight!
  [arXiv] [code] News coverage
  • This paper discovers that spurious correlations are learned from a very small fraction of the samples containing spurious features. They can be removed from the dataset even if one cannot determine/infer what spurious features/correlations are present in the dataset, to mitigate spurious correlations.
• “The Over-Certainty Phenomenon in Modern Test-Time Adaptation Algorithms”
  Fin Amin and Jung-Eun Kim
  TMLR 2025
  [arXiv]
• “RESQUE: Quantifying Estimator to Task and Distribution Shift for Sustainable Model Reusability”
  Vishwesh Sangarya and Jung-Eun Kim
  AAAI 2025
  [arXiv] [code] News coverage (Highlighted in ACM Tech News)
•  “Trustworthy AI: Safety, Bias, and Privacy – A Survey”
  Xingli Fang *, Jianwei Li *, Varun Mulchandani *, and Jung-Eun Kim,
  * Equal contribution. In alphabetical order by last name.
  CoRR abs/2502.10450, 2025
  [arXiv]
• “Superficial Safety Alignment Hypothesis”
  Jianwei Li and Jung-Eun Kim
  CoRR abs/2410.10862, 2024
  (This work is about LLM’s Safety Alignment.)
  [arXiv]
• Abhimanyu Bellam and Jung-Eun Kim, “What Causes a Disparate Impact in a Quantized Model?,” FITML at NeurIPS, 2024
• Xingli Fang and Jung-Eun Kim, “Representation Magnitude has a Liability to Privacy Vulnerability,” the AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society, 2024
  [arXiv] [code]
• Vishwesh Sangarya, Richard Bradford, and Jung-Eun Kim, “Estimating Environmental Cost Throughout Model’s Adaptive Life Cycle,” the AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society, 2024
  [arXiv] [code]
• Xingli Fang and Jung-Eun Kim, “Center-Based Relaxed Learning Against Membership Inference Attacks,” the 40th Conference on Uncertainty in Artificial Intelligence (UAI), PMLR, 2024
  [arXiv] [code]
• Divyang Doshi and Jung-Eun Kim, “ReffAKD: Resource-efficient Autoencoder-based Knowledge Distillation,” in Advancing neural network training at NeurIPS, 2023
  [arXiv] [code]
• Xingli Fang, Richard Bradford, and Jung-Eun Kim, “Cooperative Learning for Cost-Adaptive Inference,” in Advancing neural network training at NeurIPS, 2023
  [arXiv]
• Vishwesh Sangarya, Richard Bradford, and Jung-Eun Kim, “Aggregate Representation Measure for Predictive Model Reusability,” in Computational sustainability at NeurIPS, 2023
  [arXiv]
• Cuong Tran, Ferdinando Fioretto, Jung-Eun Kim, and Rakshit Naidu. “Pruning has a disparate impact on model accuracy,” in NeurIPS, 2022
  • 🏆 Spotlight, Nominated for Best Paper Award, News coverage
• Mengqi Liu, Zhong Shao, Hao Chen, Man-Ki Yoon, and Jung-Eun Kim, “Compositional Virtual Timelines: Verifying Dynamic-Priority Partitions with Algorithmic Temporal Isolation,” in Proceedings of the ACM on Programming Languages (PACMPL), Volume 6, Number OOPSLA2, Article 127, Oct. 2022
• Man-Ki Yoon, Jung-Eun Kim, Richard Bradford, Zhong Shao, “TimeDice: Schedulability-Preserving Priority Inversion for Mitigating Covert Timing Channels Between Real-time Partitions,” in Proceedings of the 52nd IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), Jun. 2022
• Man-Ki Yoon, Mengqi Liu, Hao Chen, Jung-Eun Kim, Zhong Shao, “Blinder: Partition-Oblivious Hierarchical Scheduling,” in Proceedings of the 30th USENIX Security Symposium, Aug. 2021
• Jung-Eun Kim, Richard Bradford, Max Del Giudice and Zhong Shao, “Adaptive Generative Modeling in Resource-Constrained Environments,” in Proceedings of the 24th ACM/IEEE Design, Automation, and Test in Europe (DATE), Feb. 2021.
• Jung-Eun Kim, Richard Bradford, Max Del Giudice and Zhong Shao, “Paired Training Framework for Time-Constrained Learning,” in Proceedings of the 24th ACM/IEEE Design, Automation, and Test in Europe (DATE), Feb. 2021.
• Jung-Eun Kim, Richard Bradford and Zhong Shao, “AnytimeNet: Controlling Time-Quality Tradeoffs in Deep Neural Network Architectures,” in Proceedings of the 23rd ACM/IEEE Design, Automation, and Test in Europe (DATE), Mar. 2020.
• Jung-Eun Kim, Richard Bradford, Man-Ki Yoon and Zhong Shao, “ABC: Abstract prediction Before Concreteness,” in Proceedings of the 23rd ACM/IEEE Design, Automation, and Test in Europe (DATE), Mar. 2020.
• Mengqi Liu, Lionel Rieg, Zhong Shao, Ronghui Gu, David Costanzo, Jung-Eun Kim and Man-Ki Yoon, “Virtual Timeline: A Formal Abstraction for Verifying Preemptive Schedulers with Temporal Isolation,” in Proceedings of the 47th ACM SIGPLAN Symposium on Principles of Programming Languages (POPL 2020), Jan. 2020.
• Jung-Eun Kim, Tarek Abdelzaher, Lui Sha, Amotz Bar-Noy, Reginald Hobbs and William Dron, “Decision-driven Scheduling“, The Journal Real-Time Systems, Vol. 55, Issue 3, pp 514-551, July 2019.
• Jongdeog Lee, Kelvin Marcus, Tarek Abdelzaher, Md Tanvir A. Amin, William Dron, Ramesh Govindan, Reginald Hobbs, Shaohan Hu, Amotz Bar-Noy, Jung-Eun Kim and Shuochao Yao, “Athena: Towards Decision-centric Anticipatory Sensor Information Delivery,” Journal of Sensor and Actuator Networks, Vol. 7, Issue 1, Jan. 2018.
• Tarek Abdelzaher, Tanvir Al Amin, Amotz Bar-Noy, William Dron, Ramesh Govindan, Reginald Hobbs, Shaohan Hu, Jung-Eun Kim, Shuochao Yao and Yiran Zhao, “Decision-driven Execution: A Distributed Resource Management Paradigm for the Age of IoT,” In Proc. 37th IEEE International Conference on Distributed Computing Systems (ICDCS), Atlanta, GA, June 2017.
• Jung-Eun Kim, “Timing Analysis in Existing and Emerging Cyber Physical Systems,” Ph.D. Dissertation, University of Illinois at Urbana-Champaign, Urbana, IL, May 2017.
• Jung-Eun Kim, Richard Bradford, Tarek Abdelzaher and Lui Sha, “A Schedulability Test for Software Migration on Multicore Systems,“ in Proceedings of the 20th ACM/IEEE Design, Automation, and Test in Europe (DATE 2017), Mar. 2017.
• Jung-Eun Kim, Tarek Abdelzaher, Lui Sha, Amotz Bar-Noy and Reginald Hobbs, “Sporadic Decision-centric Data Scheduling with Normally-off Sensors,” in Proceedings of the 37th IEEE Real-Time Systems Symposium (RTSS 2016), Dec. 2016.
• Jung-Eun Kim, Richard Bradford, Tarek Abdelzaher and Lui Sha, “Schedulability Analysis for Certification-friendly Multicore Systems,” Technical report, Department of Computer Science, University of Illinois at Urbana-Champaign, Nov. 2016.
• Jung-Eun Kim, Tarek Abdelzaher, Lui Sha, Amotz Bar-Noy, Reginald Hobbs and William Dron, “On Maximizing Quality of Information for the Internet of Things: A Real-time Scheduling Perspective (Invited Paper),” in Proceedings of the 22nd IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, Aug. 2016.
• Lui Sha, Marco Caccamo, Renato Mancuso, Jung-Eun Kim, Man-Ki Yoon, Rodolfo Pellizzoni, Heechul Yun, Russell Kegley, Dennis Perlman, Greg Arundale and Richard Bradford, “Real-Time Computing on Multicore Processors,” in IEEE Computer, Vol. 49, no. 9, pp. 69-77, Sep. 2016.
• Jung-Eun Kim, Tarek Abdelzaher and Lui Sha, “Budgeted Generalized Rate Monotonic Analysis for the Partitioned, yet Globally Scheduled Uniprocessor Model,” in Proceedings of the 21st IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2015 at CPS Week 2015), Apr. 2015.
• Jung-Eun Kim, Tarek Abdelzaher and Lui Sha, “Schedulability Bound for Integrated Modular Avionics Partitions,” in Proceedings of the 18th ACM/IEEE Design, Automation, and Test in Europe (DATE 2015), Mar. 2015.
• Lui Sha, Marco Caccamo, Renato Mancuso, Jung-Eun Kim, Man-Ki Yoon, Rodolfo Pellizzoni, Heechul Yun, Russel Kegley, Dennis Perlman, Greg Arundale and Richard Bradford, “Single Core Equivalent Virtual Machines for Hard Real-Time Computing on Multicore Processors,” Technical report, Department of Computer Science, University of Illinois at Urbana-Champaign, Nov. 2014.
• Jung-Eun Kim, Man-Ki Yoon, Richard Bradford and Lui Sha, “Integrated Modular Avionics (IMA) Partition Scheduling with Conflict-Free I/O for Multicore Avionics Systems,” in Proceedings of the 38th IEEE Computer Software and Applications Conference, Jul. 2014.
• Man-Ki Yoon, Sibin Mohan, Jaesik Choi, Jung-Eun Kim and Lui Sha, “SecureCore: A Multicore-based Intrusion Detection Architecture for Real-Time Embedded Systems,” in Proceedings of the 19th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2013 at CPS Week 2013), Apr. 2013.
• Jung-Eun Kim, Man-Ki Yoon, Sungjin Im, Richard Bradford and Lui Sha, “Optimized Scheduling of Multi-IMA Partitions with Exclusive Region for Synchronized Real-Time Multi-Core System,” in Proceedings of the 16th ACM/IEEE Design, Automation, and Test in Europe (DATE 2013), Mar. 2013.
• Man-Ki Yoon, Jung-Eun Kim, Richard Bradford and Lui Sha, “Holistic Design Parameter Optimization of Multiple Periodic Resources in Hierarchical Scheduling,” in Proceedings of the 16th ACM/IEEE Design, Automation, and Test in Europe (DATE 2013), Mar. 2013.
• Min-Young Nam, Kyungtae Kang, Rodolfo Pellizzoni, Kyung-Joon Park, Jung-Eun Kim and Lui Sha, “Modeling Towards Incremental Early Analyzability of Networked Avionics Systems using Virtual Integration,” in ACM Transactions on Embedded Computing Systems, Vol. 11, no. 4, pp. 81:1–81:23, Dec. 2012.
• Jung-Eun Kim, Man-Ki Yoon, Sungjin Im, Richard Bradford and Lui Sha, “Multi-IMA Partition Scheduling with Synchronized Solo-Partitions for Multi-Core Avionics Systems,” Technical report, Department of Computer Science, University of Illinois at Urbana-Champaign, May. 2012.
• Man-Ki Yoon, Jung-Eun Kim and Lui Sha, “Optimizing Tunable WCET with Shared Resource Allocation and Arbitration in Hard Real-Time Multicore Systems,” in Proceedings of the 32th IEEE Real-Time Systems Symposium (RTSS 2011), Nov. 2011.
• Man-Ki Yoon, Jung-Eun Kim and Lui Sha, “WCET-Aware Optimization of Shared Cache Partition and Bus Arbitration for Hard Real-Time Multicore Systems,” Technical report, Department of Computer Science, University of Illinois at Urbana-Champaign, http://www.ideals.illinois.edu/handle/2142/25909, May 2011.
• Jung-Eun Kim, Junghee Han and Chang-Gun Lee, “Optimal 3-Coverage with Minimum Separation Requirements for Ubiquitous Computing Environments,” ACM/Springer Mobile Networks and Applications, Vol. 14, Issue 5, pp. 556-570, Oct. 2009.
• Man-Ki Yoon, Jung-Eun Kim, Kyungtae Kang, Kyung-Joon Park, Min-Young Nam and Lui Sha, “End-to-End Delay Analysis of Wireless ECG over Cellular Networks,” in Proceedings of the 1st ACM International Workshop on Medical-grade Wireless Networks (WiMD), pp. 21-26, May. 2009.
• Jung-Eun Kim, Man-Ki Yoon, Junghee Han and Chang-Gun Lee, “Sensor Placement for 3-Coverage with Minimum Separation Requirements,” in Proceedings of the 4th IEEE International Conference on Distributed Computing in Sensor Systems (DCOSS 2008), Jun. 2008.
• Min-Young Nam, Mhd Zaher Al-Sabbagh, Jung-Eun Kim, Man-Ki Yoon, Chang-Gun Lee and Eun-Yong Ha, “A Real-time Ubiquitous System for Assisted Living: Combined Scheduling of Sensing and Communication for Real-Time Tracking,” in IEEE Transactions on Computers (TC), Vol. 57, no. 6, pp. 795-808, Jun. 2008.
• Jung-Eun Kim, Man-Ki Yoon, Junghee Han, Chang-Gun Lee and Eun Yong Ha, “Optimal Sensor Placement method for Construction of Ubiquitous Sensing Infra,” Korea Computer Congress (KCC 2008), Jun. 2008.
• Min-Young Nam, Mhd Zaher Al-Sabbagh, Jung-Eun Kim, Man-Ki Yoon, Chang-Gun Lee and Eun Yong Ha. “A Real-Time Ubiquitous System: Real-Time Indoor Tracking of Humans and Objects for Assisted Living,” Technical Report, School of Computer Science and Engineering, Seoul National University, Aug. 2007.

 

Patent

• Chang-Gun Lee, Jung-Eun Kim, and Junghee Han. Sensor Deployment System for 3-Coverage. KR 10-1032998, filed Dec. 30, 2008, and issued Apr. 27, 2011.
• Divyang Doshi and Jung-Eun Kim, “ReffAKD: Resource-efficient Autoencoder-based Knowledge Distillation,” US Patent pending