I have been a researcher at the Robotics Institute, Carnegie Mellon University (CMU), since 2019. My research spans efficient AI, embodied AI, computer vision, wireless perception, and healthcare.

Before research, I worked in industry on GPU systems and AI algorithms at NVIDIA, and on telecom embedded software, Linux systems, and distributed network programming at Motorola, Agilent Technologies, etc.

My work has appeared at NeurIPS, ICLR, CVPR, ECCV, AAAI, IJCAI, EMNLP, AAMAS, and IROS, and in IEEE TCAD and ACM TECS. I serve as a reviewer for NeurIPS, ICLR, CVPR, AAAI, AAMAS, and EMNLP.

Education

  • Ph.D.Northeastern University, Boston, USA
  • M.S. and B.S.Beihang University, Beijing, China

Honors

  • 2026IEEE Senior Member
IEEE Senior Member plaque awarded to Jun Liu, January 2026

Selected publications

  • AAMAS 2026Structured Agent Distillation for Large Language ModelsJ. Liu, Z. Kong, P. Dong, C. Yang, H. Tang, G. Yuan, W. Niu, et al.
  • Findings of EMNLP 2026RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured MemoryJ. Liu, Z. Kong, C. Yang, F. Yang, T. Li, P. Dong, J. Nanjekye, et al.
  • IROS 2026IndoorR2X: Indoor Robot-to-Everything Coordination with LLM-Driven PlanningF. Yang, S. Teotia, S. A. Mehta, J. Liu, P. Krishnakumar, Q. Xie, Y. Song, W. Li, A. Moteki, K. Uchino, Y. Bisk.
  • Findings of EMNLP 2026End-to-end on-device quantization-aware training for LLMs at inference costQ. Tan, X. Song, J. Lu, J. Liu, G. Li, L. Hong, C. Ding, J. Li, X. Zhai, S. Huang, et al.
  • CVPR 2026Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion ModelsC. Zhang, Z. Ding, C. Yang, J. Liu, X. Zhai, S. Huang, B. Li, et al.
  • NeurIPS 2025Harmony in Divergence: Towards Fast, Accurate, and Memory-efficient Zeroth-order LLM Fine-tuningQ. Tan, J. Liu, Z. Zhan, C. Ding, Y. Wang, et al.
  • ICLR 2025Mutual Effort for Efficiency: A Similarity-based Token Pruning for Vision Transformers in Self-Supervised LearningS. Li, Q. Tan, Y. Dai, J. Liu, Z. Kong, T. Wang, A. Li, N. Liu, et al.
  • AAAI 2025Toward adaptive large language models structured pruning via hybrid-grained weight importance assessmentJ. Liu, Z. Kong, P. Zhao, C. Yang, X. Shen, H. Tang, G. Yuan, et al.
  • ICASSP 2025RoRA: Efficient Fine-Tuning of LLM with Reliability Optimization for Rank AdaptationJ. Liu, Z. Kong, P. Dong, C. Yang, S. Xuan, C. Yang, P. Zhao, H. Tang, et al.
  • IJCAI 2025fairgnn-wod: Fair graph learning without complete demographicsZ. Wang, F. Liu, S. Pan, J. Liu, F. Saeed, M. Qiu, W. Zhang.
  • IJCAI 2025FairSMOE: Mitigating Multi-Attribute Fairness Problem with Sparse Mixture-of-ExpertsC. Yang, Z. Zhan, C. Zhang, J. Liu, Y. Gong, X. Shen, H. Tang, et al.
  • ACM TACO 2025Mobile-3DCNN: An Acceleration Framework for Ultra-Real-Time Execution of Large 3D CNNs on Mobile DevicesW. Niu, M. Sun, Z. Li, J. Chen, J. Liu, C. Guan, S. Xipeng, et al.
  • PLOS Digital Health 2025AI-driven healthcare: Fairness in AI healthcare: A surveyS.V. Chinta, Z. Wang, A. Palikhe, X. Zhang, A. Kashif, M.A. Smith, J. Liu*, W. Zhang. vol. 4, no. 5, e0000864. *Corresponding author.
  • IEEE TCAD 2024TSLA: A Task-Specific Learning Adaptation for Semantic Segmentation on Autonomous Vehicles PlatformJ. Liu, Z. Kong, P. Zhao, W. Zeng, H. Tang, X. Shen, C. Yang, W. Zhang, et al.
  • ECCV 2024Instructgie: Towards generalizable image editingZ. Meng, C. Yang, J. Liu, H. Tang, P. Zhao, Y. Wang.
  • ACM SIGKDD Explorations 2025Graph Fairness via Authentic Counterfactuals: Tackling Structural and Causal ChallengesZ. Wang, Z. Yin, C. Lisetti, R. Yu, S. Wang, J. Liu, S. Ganapati, et al. vol. 26, no. 2, pp. 89–98.
  • GLSVLSI 2025Towards Memory-Efficient and Sustainable Machine Unlearning on Edge using Zeroth-Order OptimizerC. Zhang, C. Yang, Q. Tan, J. Liu, A. Li, Y. Wang, J. Lu, J. Wang, G. Yuan. pp. 227–232.
  • ICCAD 2025Perturbation-efficient zeroth-order optimization for hardware-friendly on-device trainingQ. Tan, S.E. Chang, R. Xia, H. Ji, J. Liu, C. Yang, C. Zhang, Z. Zhan, Z. Zou, et al.
  • ASP-DAC 2025A computation and energy efficient hardware architecture for SSL accelerationH. Ji, S. Li, Y. Cao, C. Ding, J. Liu, J. Xu, Q. Tan, A. Li, X. Tang, L. Zheng, et al. pp. 23–29.
  • ICASSP 2024Df-vton: Dense flow guided virtual try-on networkH. Dong, J. Liu*, D. Huang. *Corresponding author.
  • WACV 2024Physical-space multi-body mesh detection achieved by local alignment and global dense learningH. Dong, T. Xiang, S. Chittupalli, J. Liu, D. Huang.
  • ACMM 2023 WorkshopA scalable real-time semantic segmentation network for autonomous drivingJ. Liu, C. Wu, G. Yuan, W. Niu, W. Zhang, H.H. Song. Workshop on Advanced Multimedia Computing for Smart Manufacturing and Engineering.
  • ICRA 2024Mono Wi-Fi SceneS. Chittupalli, J. Liu, D. Huang.
  • ICRA 2024Universal Monocular 3D Human Recovery EngineH. Dong, J. Liu, D. Huang.
  • ACM TECS 2022Mobile or FPGA? A comprehensive evaluation on energy efficiency and a unified optimization frameworkG. Yuan, P. Dong, M. Sun, W. Niu, J. Liu, Z. Li, Y. Cai, Y. Li, W. Jiang, X. Lin, et al.
  • IEEE RTAS 2021Work in progress: Mobile or FPGA? A comprehensive evaluation on energy efficiency and a unified optimization frameworkG. Yuan, P. Dong, M. Sun, W. Niu, Z. Li, Y. Cai, J. Liu, W. Jiang, X. Lin, B. Ren, et al.

Patents