70 citations · 127 across the 6 of their papers we have counts for
6 papers
AInfer-PD: Communication-Safe In-Place Prefill-Decode Multiplexing for Distributed MoE Rollouts
Guowei Wang, Chaokun Yang, Zhenxuan Pan +6
Rollout inference often dominates the wall-clock time of large-scale reinforcement learning (RL). In agentic RL, each trajectory alternates between model generation and environment…
HARTS: Efficient Agentic Reinforcement Learning for Hybrid-Attention Models over Arbitrary Rollout Trees
Boyuan Meng, Peihua Bao, Hong Liu +4
Agentic reinforcement learning (RL) often produces irregular rollout trees with shared histories. Training root-to-leaf trajectories independently recomputes these shared prefixes.…
RoutePack: Replay-Guided Expert Placement and Rollout Trajectory Scheduling for MoE Reinforcement Learning
Yibo Shen, Xudong Han, Xiaowei Zhu +2
Reinforcement learning (RL) post-training of Mixture-of-Experts (MoE) models couples two load-balancing problems. Rollout-trajectory composition determines attention and dense feed…
DFOGraph: An I/O- and Communication-Efficient System for Distributed Fully-out-of-Core Graph Processing
Jiping Yu, Wei Qin, Xiaowei Zhu +4
With the magnitude of graph-structured data continually increasing, graph processing systems that can scale-out and scale-up are needed to handle extreme-scale datasets. While exis…
RisGraph: A Real-Time Streaming System for Evolving Graphs to Support Sub-millisecond Per-update Analysis at Millions Ops/s
Guanyu Feng, Zixuan Ma, Daixuan Li +4
Evolving graphs in the real world are large-scale and constantly changing, as hundreds of thousands of updates may come every second. Monotonic algorithms such as Reachability and…
LiveGraph: A Transactional Graph Storage System with Purely Sequential Adjacency List Scans
Xiaowei Zhu, Guanyu Feng, Marco Serafini +5
The specific characteristics of graph workloads make it hard to design a one-size-fits-all graph storage system. Systems that support transactional updates use data structures with…