publications

2026

1. preprint

   TStore: Rethinking AI Model Hub with Tensor-Centric Compression

   Tingfeng Lan, Zirui Wang, Yunjia Zheng, and 3 more authors

   2026

   Link
2. NSDI’26

   ZipLLM: Efficient LLM Storage via Model-Aware Synergistic Data Deduplication and Compression

   Zirui Wang, Tingfeng Lan, Zhaoyuan Su, and 2 more authors

   2026

   Link
3. MLSys’26

   MorphServe: Efficient and Workload-Aware LLM Serving via Runtime Layer Swapping and KV Cache Resizing

   Zhaoyuan Su, Tingfeng Lan, Zirui Wang, and 2 more authors

   In Proceedings of Machine Learning and Systems, 2026

   Link Bib

   @inproceedings{su2025morphserve,
     title = {MorphServe: Efficient and Workload-Aware LLM Serving via Runtime Layer Swapping and KV Cache Resizing},
     author = {Su, Zhaoyuan and Lan, Tingfeng and Wang, Zirui and Yang, Juncheng and Cheng, Yue},
     booktitle = {Proceedings of Machine Learning and Systems},
     year = {2026},
     eprint = {2506.02006},
     archiveprefix = {arXiv},
     primaryclass = {cs.DC},
     url = {https://arxiv.org/abs/2506.02006},
   }

4. MLSys’26

   λScale: Enabling Fast Scaling for Serverless Large Language Model Inference

   Minchen Yu, Rui Yang, Chaobo Jia, and 8 more authors

   In Proceedings of Machine Learning and Systems, 2026

   Link Bib

   @inproceedings{yu2025lambdascale,
     title = {{\lambda}Scale: Enabling Fast Scaling for Serverless Large Language Model Inference},
     author = {Yu, Minchen and Yang, Rui and Jia, Chaobo and Su, Zhaoyuan and Yao, Sheng and Lan, Tingfeng and Yang, Yuchen and Cheng, Yue and Wang, Wei and Wang, Ao and Chen, Ruichuan},
     booktitle = {Proceedings of Machine Learning and Systems},
     year = {2026},
     eprint = {2502.09922},
     archiveprefix = {arXiv},
     primaryclass = {cs.DC},
     url = {https://arxiv.org/abs/2502.09922},
   }

2025

1. preprint

   ZenFlow: Enabling Stall-Free Offloading Training via Asynchronous Updates

   Tingfeng Lan, Yusen Wu, Bin Ma, and 7 more authors

   2025

   Link
2. VLDB’25

   mLoRA: Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs

   Zhengmao Ye, Dengchun Li, Zetao Hu, and 8 more authors

   2025

   Link

2024

1. VLDB’24

   DLRover-RM: Resource Optimization for Deep Recommendation Models Training in the Cloud

   Qinlong Wang^*, Tingfeng Lan^*, Yinghao Tang, and 8 more authors

   Proceedings of the VLDB Endowment, 2024