3 papers
cs.LG2026
JigsawRL: Assembling RL Pipelines for Efficient LLM Post-Training
Zhengding Hu, Hehua Ouyang, Chang Chen +6
We present JigsawRL, a cost-efficient framework that explores Pipeline Multiplexing as a new dimension of RL parallelism. JigsawRL decomposes each pipeline into a Sub-Stage Graph t…
cs.MA2026
Pancake: Hierarchical Memory System for Multi-Agent LLM Serving
Zhengding Hu, Zaifeng Pan, Prabhleen Kaur +6
In this work, we identify and address the core challenges of agentic memory management in LLM serving, where large-scale storage, frequent updates, and multiple coexisting agents j…
cs.PF2024
The Hitchhiker's Guide to Programming and Optimizing Cache Coherent Heterogeneous Systems: CXL, NVLink-C2C, and AMD Infinity Fabric
Zixuan Wang, Suyash Mahar, Luyi Li +9
We present a thorough analysis of the use of modern heterogeneous systems interconnected by various cachecoherent links, including CXL, NVLink-C2C, and Infinity Fabric. We studied…