February 04, 2026
Xinyue Ma is accepted to the intern program at Microsoft Research Redmond
Xinyue Ma is accepted to the intern program at Microsoft Research Redmond
Internship
Taeyoon Kim is accepted to the intern program at IBM Reaserch Tokyo
Internship
Xinyue Ma is accepted to the intern program at Microsoft Research Redmond
VLDB 2026
ORBITFLOW: SLO-Aware Long-Context LLM Serving with Fine-Grained KV Cache Reconfiguration
NeurIPS 2025
Sungho Jeon participated NeurIPS 2025 Conference
Eurosys 2026
Garen: Relieable Cluster Management with Atomic State Reconcilation
Eurosys 2026
CCL: Carbon-Aware Continuous Learning for Sustainable Real-Time Machine Learning Analytics
Our recent publications / View more
VLDB
ORBITFLOW: SLO-Aware Long-Context LLM Serving with Fine-Grained KV Cache Reconfiguration more
Xinyue Ma*, Heelim Hong*, Jongseop Lee, Seoyeong Choy, Woo-Yeon Lee, Taegeon Um, Myeongjae Jeon (*Equal Contribution)
In Proceedings of the 52nd International Conference on Very Large Data Bases. Aug. 2026.
EuroSys
Carbon-Aware Continuous Learning for Sustainable Real-Time Machine Learning Analytics more
Gwanjong Park, Osama Khan, Dongho Ha, Myeongjae Jeon, Euiseong Seo
In Proceedings of the 21th European Conference on Computer Systems. Apr. 2026.
EuroSys
Garen: Reliable Cluster Management with Atomic State Reconciliation more
Mingi Kim, Ahnjae Shin, Jaewoo Maeng, Myeongjae Jeon, Byung-Gon Chun
In Proceedings of the 21th European Conference on Computer Systems. Apr. 2026.
Our research / View more
Efficient LLM Serving Systems
Collaborators: ETRI, Microsoft Research Redmond, Samsung Research, KAIST, Samsung DS
- SLO-aware, adaptive KV cache management
- Generalized KV cache reuse
- Accuracy-preserving long-context pruning
- SLO-aware failure recovery for LLMs
- Systems for agentic AI
Continual and On-Device Learning
Collaborators: Palantir, SNU, UIUC
- Low-latency, high-accuracy on-device learning
- Adaptive, resource-efficient continual learning
Large-Scale Distributed Training
Collaborators: Samsung Research, KAIST, UC Merced, USC
- Learning-based planning for heterogeneous, geo-distributed training
- Fast distributed training on heterogeneous accelerators
Fast and Scalable Big Data Analytics
Collaborators: Amazon, Samsung Electronics, SNU
- Efficient caching for iterative analytics
- Data preprocessing for scalable ML pipelines
Our Collaborators
Our recent posts / View more posts
February 03, 2026
Taeyoon Kim is accepted to the intern program at IBM Research Tokyo
Taeyoon Kim is accepted to the intern program at IBM Research Tokyo
December 16, 2025
ORBITFLOW is accepted to VLDB 2026
ORBITFLOW is accepted to the VLDB 2026, 52nd International Conference on Very Large Data Bases.
