arXiv subjects

Jianmin Hu

Publications and source records attributed to Jianmin Hu.

6 recordsLinked to original sources

arXiv2026

RolloutPipe: Overlapping Pipelined Rollout and Training in Disaggregated On-Policy LLM Reinforcement Learning

Large language model (LLM) post-training for reasoning increasingly relies on reinforcement learning with verifiable rewards (RLVR), where models learn from ground-truth feedback on mathematical, logical, and scientific tasks. To enable flexible resource allocation and support heterogeneous training setups, modern RLVR systems adopt disaggregated architectures that decouple rollout generation and policy training across independent GPU pools. However, existing synchronous on-policy GRPO (Group Relative Policy Optimization) RLVR systems finish an entire rollout before starting training, leaving the trainer GPU pool idle while rollout is still ongoing. Asynchronous RL pipelines overlap the two stages, but at the cost of training on stale data. To address these challenges, we propose RolloutPipe, a post-training framework for disaggregated RLVR systems, which turns the fixed-weight rollout into a complete-group pipeline where trainable groups move to the trainer while later groups are still being generated. RolloutPipe achieves this through two techniques including complete-group pipelining (CGP) and frontier-group dispatch (FGD). CGP dispatches each trainable complete group to the trainer FIFO as soon as group materialization finishes, and FGD is an admission policy on the Rollout node that first admits requests for the frontier groups needed to form the next training batch, so that trainer-ready groups arrive earlier and more steadily. The design starts training before the rollout completes while maintaining on-policy correctness. Evaluated on Qwen3-1.7B across four reasoning and science benchmarks and twelve rollout settings, RolloutPipe shortens the rollout-to-train-end time by 30.7%-42.3%, and lowers the trainer waiting ratio by 37%-76% compared to Slime, a state-of-the-art rollout and training system.

cs.DC↗

arXiv2026

SwiftCache: Efficient LLM Serving for Multi-turn Conversations with Heterogeneous KV Cache Sharing

Multi-turn conversation is a fundamental scenario in LLM applications, widely used in chatbots and AI agents. As the conversation evolves, historical tokens accumulate continuously. Existing systems cache their key-value (KV) pairs to avoid redundant computation. However, limited GPU memory (HBM) capacity often forces these KV caches to be offloaded to CPU memory or SSD, making KV cache reloads increasingly costly in terms of latency as the context grows. Meanwhile, the constrained HBM capacity also limits the maximum inference length, thereby restricting the number of turns that can be supported in a conversation. To address these two challenges, we propose SwiftCache, a collaborative inference system that enables heterogeneous models to share underutilized GPU memory and NVLink bandwidth within a server. Specifically, models with low KV cache demand donate idle GPU memory to store the prefix cache of high-demand models, allowing cross-model KV cache sharing over NVLink and avoiding slow PCIe transfers. SwiftCache further reduces memory pressure by keeping only the KV cache of the currently active layer in local GPU memory, thereby enabling longer-context inference. Our experiments on real-world workloads show that SwiftCache reduces P99 time-to-first-token (TTFT) by up to 69% and extends maximum context length by up to 3.98x compared to vLLM and SGLang, with minimal interference to co-located models.

cs.DC↗

arXiv2025

End-to-End Motion Capture from Rigid Body Markers with Geodesic Loss

Marker-based optical motion capture (MoCap), while long regarded as the gold standard for accuracy, faces practical challenges, such as time-consuming preparation and marker identification ambiguity, due to its reliance on dense marker configurations, which fundamentally limit its scalability. To address this, we introduce a novel fundamental unit for MoCap, the Rigid Body Marker (RBM), which provides unambiguous 6-DoF data and drastically simplifies setup. Leveraging this new data modality, we develop a deep-learning-based regression model that directly estimates SMPL parameters under a geodesic loss. This end-to-end approach matches the performance of optimization-based methods while requiring over an order of magnitude less computation. Trained on synthesized data from the AMASS dataset, our end-to-end model achieves state-of-the-art accuracy in body pose estimation. Real-world data captured using a Vicon optical tracking system further demonstrates the practical viability of our approach. Overall, the results show that combining sparse 6-DoF RBM with a manifold-aware geodesic loss yields a practical and high-fidelity solution for real-time MoCap in graphics, virtual reality, and biomechanics.

cs.CV↗

arXiv2025

BanaServe: Unified KV Cache and Dynamic Module Migration for Balancing Disaggregated LLM Serving in AI Infrastructure

Large language models (LLMs) are increasingly deployed in AI infrastructure, driving the need for high throughput, resource efficient serving systems. Disaggregated LLM serving, which separates prompt prefill from auto-regressive decode, has emerged as a promising architecture by isolating their heterogeneous compute and memory demands. However, current disaggregated systems face three key limitations: (i) static resource allocation cannot adapt to highly dynamic workloads, causing over-provisioning that wastes resources or under-provisioning that violates service level objectives (SLOs); (ii) inherent load imbalance between prefill and decode stages, where prefill is compute-bound and decode is memory-bound, causes under-utilization in one tier while the other becomes a bottleneck; and (iii) prefix cache aware routing skews load distribution, as high cache hit rate prefill nodes attract disproportionately more requests, further degrading balance and efficiency. To address these issues, we present BanaServe, a dynamic orchestration framework that continuously rebalances computational and memory resources across prefill and decode instances while eliminating hotspots induced by cache. BanaServe introduces layer level weight migration, attention level Key Value Cache (KV Cache) migration, and Global KV Cache Store sharing with layer wise overlapped transmission, enabling both coarse grained (layer level) and fine grained (attention level) load redistribution with minimal latency overhead. These mechanisms allow routers to perform purely load aware scheduling, unconstrained by cache placement. Compared to vLLM, BanaServe achieves 1.2x-3.9x higher throughput with 3.9%-78.4% lower total processing time, and outperforms DistServe by 1.1x-2.8x in throughput with 1.4%-70.1% latency reduction.

cs.DC↗

arXiv2025

BrownoutServe: SLO-Aware Inference Serving under Bursty Workloads for MoE-based LLMs

In recent years, the Mixture-of-Experts (MoE) architecture has been widely applied to large language models (LLMs), providing a promising solution that activates only a subset of the model's parameters during computation, thereby reducing overall memory requirements and allowing for faster inference compared to dense models. Despite these advantages, existing systems still face issues of low efficiency due to static model placement and lack of dynamic workloads adaptation. This leads to suboptimal resource utilization and increased latency, especially during bursty requests periods. To address these challenges, this paper introduces BrownoutServe, a novel serving framework designed to optimize inference efficiency and maintain service reliability for MoE-based LLMs under dynamic computational demands and traffic conditions. BrownoutServe introduces "united experts" that integrate knowledge from multiple experts, reducing the times of expert access and inference latency. Additionally, it proposes a dynamic brownout mechanism to adaptively adjust the processing of certain tokens, optimizing inference performance while guaranteeing service level objectives (SLOs) are met. Our evaluations show the effectiveness of BrownoutServe under various workloads: it achieves up to 2.07x throughput improvement compared to vLLM and reduces SLO violations by 90.28%, showcasing its robustness under bursty traffic while maintaining acceptable inference accuracy.

cs.DC↗

arXiv2024

Edge AI: A Taxonomy, Systematic Review and Future Directions

Edge Artificial Intelligence (AI) incorporates a network of interconnected systems and devices that receive, cache, process, and analyze data in close communication with the location where the data is captured with AI technology. Recent advancements in AI efficiency, the widespread use of Internet of Things (IoT) devices, and the emergence of edge computing have unlocked the enormous scope of Edge AI. Edge AI aims to optimize data processing efficiency and velocity while ensuring data confidentiality and integrity. Despite being a relatively new field of research from 2014 to the present, it has shown significant and rapid development over the last five years. This article presents a systematic literature review for Edge AI to discuss the existing research, recent advancements, and future research directions. We created a collaborative edge AI learning system for cloud and edge computing analysis, including an in-depth study of the architectures that facilitate this mechanism. The taxonomy for Edge AI facilitates the classification and configuration of Edge AI systems while examining its potential influence across many fields through compassing infrastructure, cloud computing, fog computing, services, use cases, ML and deep learning, and resource management. This study highlights the significance of Edge AI in processing real-time data at the edge of the network. Additionally, it emphasizes the research challenges encountered by Edge AI systems, including constraints on resources, vulnerabilities to security threats, and problems with scalability. Finally, this study highlights the potential future research directions that aim to address the current limitations of Edge AI by providing innovative solutions.

cs.DC↗