arXiv subjects

Qingyu Ma

Publications and source records attributed to Qingyu Ma.

5 recordsLinked to original sources

arXiv2026

ATOM: Instantiating Budget-Controllable Multi-Agent Collaboration via Nucleus-Electron Hierarchy

Large Language Model (LLM)-based multi-agent systems rely on optimized collaboration topologies to balance performance and communication costs. However, current methods struggle with the inherent stability-extensibility trade-off and often misalign computational budgets with query difficulty. We propose \textsc{ATOM}, an adaptive framework that generates budget-controllable collaboration graphs via a novel task-driven reinforcement learning paradigm. Inspired by atomic structures, \textsc{ATOM} employs a nucleus-electron hierarchy: it maintains a stable, offline-learned collaboration backbone (the nucleus) while dynamically activating query-conditioned agents (electrons) during inference. Crucially, a complexity-aware budgeting strategy aligns resource consumption with task demands by estimating query difficulty to strictly regulate electron instantiation. Extensive experiments across six diverse benchmarks demonstrate that \textsc{ATOM} achieves state-of-the-art performance while improving token efficiency by up to $30\%$ compared to strong baselines.

cs.MA↗

arXiv2026

ProMAS: Proactive Error Forecasting for Multi-Agent Systems Using Markov Transition Dynamics

The integration of Large Language Models into Multi-Agent Systems (MAS) has enabled the so-lution of complex, long-horizon tasks through collaborative reasoning. However, this collec-tive intelligence is inherently fragile, as a single logical fallacy can rapidly propagate and lead to system-wide failure. Most current research re-lies on post-hoc failure analysis, thereby hinder-ing real-time intervention. To address this, we propose PROMAS, a proactive framework utiliz-ing Markov transitions for predictive error anal-ysis. PROMAS extracts Causal Delta Features to capture semantic displacement, mapping them to a quantized Vector Markov Space to model reasoning as probabilistic transitions. By inte-grating a Proactive Prediction Head with Jump Detection, the method localizes errors via risk acceleration rather than static thresholds. On the Who&When benchmark, PROMAS achieves 22.97% step-level accuracy while processing only 27% of reasoning logs. This performance rivals reactive monitors like MASC while reducing data overhead by 73%. Although this strategy entails an accuracy trade-off compared to post-hoc meth-ods, it significantly improves intervention latency, balancing diagnostic precision with the real-time demands of autonomous reasoning.

cs.AI↗

arXiv2025

TenonOS: A Self-Generating LibOS-on-LibOS Framework for Time-Critical Embedded Operating Systems

The growing complexity of embedded systems creates tension between rich functionality and strict resource and real-time constraints. Traditional monolithic operating system and hypervisor designs suffer from resource bloat and unpredictable scheduling, making them unsuitable for time-critical workloads where low latency and low jitter are essential. We propose TenonOS, a demand-driven, self-generating, lightweight operating system framework for time-critical embedded systems that rethinks both hypervisor and operating system architectures. TenonOS introduces a LibOS-on-LibOS model that decomposes hypervisor and operating system functionality into fine-grained, reusable micro-libraries. A generative orchestration engine dynamically composes these libraries to synthesize a customized runtime tailored to each application's criticality, timing requirements, and resource profile. TenonOS consists of two core components: Mortise, a minimalist micro-hypervisor, and Tenon, a real-time library operating system. Mortise provides lightweight isolation and removes the usual double-scheduler overhead in virtualized setups, while Tenon provides precise and deterministic task management. By generating only the necessary software stack per workload, TenonOS removes redundant layers, minimizes the trusted computing base, and maximizes responsiveness. Experiments show a 40.28 percent reduction in scheduling latency, an ultra-compact 361 KiB memory footprint, and strong adaptability.

cs.OS↗

arXiv2025

MUSE: A Heterogeneity-Aware Multimedia Search Engine for Mobile SoCs

On-device multimedia retrieval is vital for smartphones, enabling applications like cross-modal semantic search and multimodal personal AI agents. However, realizing efficient retrieval on mobile SoCs remains a critical systems challenge. Unlike text-based search, multimedia applications depend on high-dimensional cross-modal embeddings (often exceeding 1024 dimensions) and continuously expanding media streams (e.g., video lifelogs and ambient audio). Mobile devices must process these intensive workloads under strict latency, energy, and memory constraints. Existing vector retrieval systems, designed primarily for static server datasets, are poorly suited for the dynamic execution models of mobile SoCs, where interactive queries must coexist with continuous background media ingestion and index maintenance. To address these challenges, we propose MUSE, a MUltimedia Search Engine comprehensively co-designed for mobile SoCs. MUSE tackles these issues through an architecture-index co-design. It introduces a three-stage asynchronous execution pipeline that overlaps DMA transfers, NPU matrix computation, and hardware-vector data adaptation to overcome the memory wall of high-dimensional embeddings. Furthermore, it redesigns the Inverted File Index (IVF) and introduces workload-aware heterogeneous scheduling across the CPU, GPU, and NPU for hybrid interactive-and-ingestion workloads. Evaluated on Snapdragon 8-series SoCs using real-world multimodal datasets, MUSE improves query throughput by up to 1.4x at matched recall, achieves up to 7x faster index construction, and delivers up to 6x higher insertion throughput under concurrent streaming. Crucially, its accelerator-native design ensures strict physical compliance, capping peak device temperatures at 38{\deg}C and reducing total energy consumption by up to 4.6x compared to CPU-bound baselines.

cs.DC↗

arXiv2018

Topological-charge dependence of radiation torque for an acoustic-vortex spanner

Based on the analyses of the wave front and the wave vector of an acoustic-vortex (AV) spanner generated by a circular array of point source, the principle of object rotation is investigated through the calculation of the orbital angular momentum (OAM) and the radiation torque for AVs with various topological charges. It is demonstrated theoretically that the rotation of the axisymmetric disk centered on an AV spanner is mainly driven by the exerted radiation torque without the contribution of the OAM transfer. The radiation torque on a small-radius object is inversely associated with the topological charge in the center of the AV spanner, and it is enhanced significantly for a larger AV with a higher topological charge. The special case of the radiation torque proportional to the topological charge might be realized when the disk radius is much larger than the wavelength, in which case the acoustic power of the AV spanner can be absorbed as much as possible. With the established sixteen-source experimental setup, the radial pressure distributions of AVs with different topological charges measured at the frequency of 1.3 kHz in air agree well with the simulations. The topological-charge dependence of the radiation torque for AVs is also verified by the quantitative laser-displacement (angle) measurements for disks with different radii. The favorable results demonstrate that, for the object covering the vortex center of an AV spanner, the total OAM transfer might be 0 and the object rotation is contributed by the acting moments. Thus, the radiation torque of an AV spanner is more applicable than the OAM for describing the driving capability of object rotation, and it can be used as an effective tool in clinical applications to noninvasively manipulate objects with a feature size at the wavelength-scale (e.g. kidney stone in lithotripsy) inside body.

physics.app-ph↗