arXiv subjects

Changyu Li

Publications and source records attributed to Changyu Li.

12 recordsLinked to original sources

arXiv2026

Beyond the All-in-One Agent: Benchmarking Role-Specialized Multi-Agent Collaboration in Enterprise Workflows

Large language model (LLM) agents are increasingly expected to operate in enterprise environments, where work is distributed across specialized roles, permission-controlled systems, and cross-departmental procedures. However, existing enterprise benchmarks largely evaluate single agents with broad tool access, while existing multi-agent benchmarks rarely capture realistic enterprise constraints such as role specialization, access control, stateful business systems, and policy-based approvals. We introduce \textsc{EntCollabBench}, a benchmark for evaluating enterprise multi-agent collaboration. \textsc{EntCollabBench} simulates a permission-isolated organization with 11 role-specialized agents across six departments and contains two evaluation subsets: a Workflow subset, where agents collaboratively modify enterprise system states, and an Approval subset, where agents make policy-grounded decisions. Evaluation is based on execution traces, database state verification, and deterministic policy adjudication rather than natural-language response judging. Experiments with representative LLM agents show that current models still struggle with end-to-end enterprise collaboration, especially in delegation, context transfer, parameter grounding, workflow closure, and decision commitment. \textsc{EntCollabBench} provides a reproducible testbed for measuring and improving agent systems intended for realistic organizational environments.

cs.MA↗

arXiv2026

FED-FSTQ: Fisher-Guided Token Quantization for Communication-Efficient Federated Fine-Tuning of LLMs on Edge Devices

Federated fine-tuning provides a practical route to adapt large language models (LLMs) on edge devices without centralizing private data. However, in mobile deployments, the training wall-clock is often dominated by straggler-limited uplink communication under heterogeneous bandwidth, intermittent participation, and non-IID client data. Although parameter-efficient fine-tuning (PEFT) methods such as LoRA and QLoRA reduce local memory and trainable parameters, repeated transmission of adapter updates remains a major bottleneck. We propose Fed-FSTQ, a semantic-sensitivity-aware communication-control primitive for communication-efficient federated LLM fine-tuning. Fed-FSTQ uses a lightweight token-level Fisher proxy to estimate semantic sensitivity, couples token-guided sparsification with mixed-precision adapter-update quantization, and allocates higher communication fidelity to semantically load-bearing evidence while suppressing redundant transmission. The method is drop-in compatible with standard federated PEFT pipelines and requires no change to the server aggregation rule. Experiments on multilingual QA and medical QA under non-IID partitions show that Fed-FSTQ reduces cumulative uplink traffic required to reach a fixed quality threshold by 46-fold relative to a Fed-LoRA baseline and improves straggler-limited wall-clock time-to-accuracy by 52%. Under the corrected Controlled LTE-20Mbps accounting, Fed-FSTQ reduces per-round time from 414.60s to 67.29s and reduces per-round energy from 839.20J to 146.28J, yielding a 6.16-fold speedup. On NVIDIA Jetson-class edge devices, Fisher-guided token reduction also yields up to a 1.55-fold inference speedup, demonstrating deployability under tight resource constraints.

cs.LG↗

arXiv2026

PI-TTA: Physics-Informed Source-Free Test-Time Adaptation for Robust Human Activity Recognition on Mobile Devices

Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device personalization from unlabeled test streams without centralizing private data. However, sensor-based human activity recognition (HAR) poses challenges that are less pronounced in standard vision benchmarks: behavioral inertial streams are temporally correlated and often exhibit within-session shifts caused by sensor rotation, placement change, and sampling-rate drift. Under this streaming non-i.i.d. setting, widely used vision-style TTA objectives can become unstable, leading to overconfident errors, representation collapse, and catastrophic forgetting. We propose PI-TTA, a lightweight source-free adaptation framework that stabilizes online updates through three physics-consistent constraints: gravity consistency, short-horizon temporal continuity, and spectral stability. PI-TTA updates the same small parameter subset as strong source-free baselines and incurs only modest overhead, making it suitable for on-device deployment. Experiments on USCHAD, PAMAP2, and mHealth under long-sequence stress tests and factorized shift protocols show that PI-TTA mitigates the severe degradation observed in confidence-driven baselines and preserves stable adaptation under sustained streaming conditions. It improves long-sequence accuracy by up to 9.13% and reduces physical-violation rates by 27.5%, 24.1%, and 45.4% on USCHAD, PAMAP2, and mHealth, respectively. These results demonstrate that physics-informed adaptation can improve accuracy, stability, and deployment reliability for real-world mobile sensing systems.

cs.AI↗

arXiv2026

Reliability-Calibrated Edge-IoT Early Fault Warning for Rotating Machinery with a Physics-Guided Tiny-Mamba Transformer

Industrial Internet of Things (IIoT) systems increasingly rely on distributed vibration sensing to support predictive maintenance of rotating machinery. In practical deployments, however, raw signal upload is costly and alarm decisions must be made locally under limited computation, changing operating conditions, and strict nuisance-alarm budgets. This paper presents a reliability-calibrated edge-IoT early-warning framework, in which a compact Physics-Guided Tiny-Mamba Transformer (PG-TMT) acts as the representation module and an extreme value theory (EVT) layer converts streaming anomaly scores into event-level alarm episodes. PG-TMT combines a depthwise-separable convolutional stem, a Tiny-Mamba state-space branch, and a lightweight local Transformer to capture transient, long-horizon, and multichannel degradation cues under batch-size-one inference. To improve auditability, temporal attention is projected to the frequency domain and softly aligned with analytical bearing fault-order bands. EVT calibration, dual-threshold hysteresis, and trimmed-tail fitting provide controllable false-alarm intensity even when healthy calibration data are imperfect. Experiments on CWRU, Paderborn, XJTU-SY, and an industrial pilot demonstrate that the proposed framework improves PR-AUC, reduces detection delay under a controlled nuisance-alarm budget, and remains robust to structured interference, metadata uncertainty, compound fault mixtures, and domain transfer. With a sub-1 MB footprint and Jetson p99 latency below 7 ms, the framework supports calibrated and interpretable early warnings for IIoT predictive maintenance.

cs.LG↗

arXiv2025

Exploring Linear Attention Alternative for Single Image Super-Resolution

Deep learning-based single-image super-resolution (SISR) technology focuses on enhancing low-resolution (LR) images into high-resolution (HR) ones. Although significant progress has been made, challenges remain in computational complexity and quality, particularly in remote sensing image processing. To address these issues, we propose our Omni-Scale RWKV Super-Resolution (OmniRWKVSR) model which presents a novel approach that combines the Receptance Weighted Key Value (RWKV) architecture with feature extraction techniques such as Visual RWKV Spatial Mixing (VRSM) and Visual RWKV Channel Mixing (VRCM), aiming to overcome the limitations of existing methods and achieve superior SISR performance. This work has proved able to provide effective solutions for high-quality image reconstruction. Under the 4x Super-Resolution tasks, compared to the MambaIR model, we achieved an average improvement of 0.26% in PSNR and 0.16% in SSIM.

cs.CV↗

arXiv2023

W-MAE: Pre-trained weather model with masked autoencoder for multi-variable weather forecasting

Weather forecasting is a long-standing computational challenge with direct societal and economic impacts. This task involves a large amount of continuous data collection and exhibits rich spatiotemporal dependencies over long periods, making it highly suitable for deep learning models. In this paper, we apply pre-training techniques to weather forecasting and propose W-MAE, a Weather model with Masked AutoEncoder pre-training for weather forecasting. W-MAE is pre-trained in a self-supervised manner to reconstruct spatial correlations within meteorological variables. On the temporal scale, we fine-tune the pre-trained W-MAE to predict the future states of meteorological variables, thereby modeling the temporal dependencies present in weather data. We conduct our experiments using the fifth-generation ECMWF Reanalysis (ERA5) data, with samples selected every six hours. Experimental results show that our W-MAE framework offers three key benefits: 1) when predicting the future state of meteorological variables, the utilization of our pre-trained W-MAE can effectively alleviate the problem of cumulative errors in prediction, maintaining stable performance in the short-to-medium term; 2) when predicting diagnostic variables (e.g., total precipitation), our model exhibits significant performance advantages over FourCastNet; 3) Our task-agnostic pre-training schema can be easily integrated with various task-specific models. When our pre-training framework is applied to FourCastNet, it yields an average 20% performance improvement in Anomaly Correlation Coefficient (ACC).

cs.LG↗

arXiv2022

Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendations

Cold-start issues have been more and more challenging for providing accurate recommendations with the fast increase of users and items. Most existing approaches attempt to solve the intractable problems via content-aware recommendations based on auxiliary information and/or cross-domain recommendations with transfer learning. Their performances are often constrained by the extremely sparse user-item interactions, unavailable side information, or very limited domain-shared users. Recently, meta-learners with meta-augmentation by adding noises to labels have been proven to be effective to avoid overfitting and shown good performance on new tasks. Motivated by the idea of meta-augmentation, in this paper, by treating a user's preference over items as a task, we propose a so-called Diverse Preference Augmentation framework with multiple source domains based on meta-learning (referred to as MetaDPA) to i) generate diverse ratings in a new domain of interest (known as target domain) to handle overfitting on the case of sparse interactions, and to ii) learn a preference model in the target domain via a meta-learning scheme to alleviate cold-start issues. Specifically, we first conduct multi-source domain adaptation by dual conditional variational autoencoders and impose a Multi-domain InfoMax (MDI) constraint on the latent representations to learn domain-shared and domain-specific preference properties. To avoid overfitting, we add a Mutually-Exclusive (ME) constraint on the output of decoders to generate diverse ratings given content data. Finally, these generated diverse ratings and the original ratings are introduced into the meta-training procedure to learn a preference meta-learner, which produces good generalization ability on cold-start recommendation tasks. Experiments on real-world datasets show our proposed MetaDPA clearly outperforms the current state-of-the-art baselines.

cs.IR↗

arXiv2019

Studies on ion back-flow of Time Projection Chamber based on GEM and anode wire grid

Gated wires are widely used in Time Projection Chamber (TPC) to avoid ion back-flow (IBF) in the drift volume. The anode wires can provide stable gain at high voltage with a long lifetime. However, switching on and off the gated grid (GG) leads to a dead time and also limit the readout efficiency of the TPC. Gas Electron Multiplier (GEM) foil provides a possibility of continuous readout for TPC, which can suppress IBF efficiently while keeping stable gain. A prototype chamber including two layers of GEM foils and anode wires has been built to combine both advantages from GEM and anode wire. Using Garfield++ and the finite element analysis (FEA) method, simulations of the transmission processes of electrons and ions are performed and results on absorption ratio of ions, gain and IBF ratio are obtained. The optimized parameters from simulation are then applied to the prototype chamber to test the IBF and other performances. Both GEM foils are run at low voltage (255V), while most of the gain is provided by the anode wire. The measurement shows that the IBF ratio can be suppressed to ~0.58% with double-layer GEM foils (staggered) at an effective gain about 2500 with an energy resolution about 10%.

physics.ins-det↗

arXiv2018

Cosmic test of sTGC detector prototype made in China for ATLAS experiment upgrade

Following the Higgs particle discovery, the Large Hadron Collider complex will be upgraded in several phases allowing the luminosity to increase to $7 \times 10^{34}cm^{-2}s^{-1}$. In order to adapt the ATLAS detector to the higher luminosity environment after the upgrade, part of the ATLAS muon end-cap system, the Small Wheel, will be replaced by the New Small Wheel. The New Small Wheel includes two kinds of detectors: small-strip Thin Gap Chambers and Micromegas. Shandong University, part of the ATLAS collaboration, participates in the construction of the ATLAS New Small Wheel by developing, producing and testing the performance of part of the small-strip Thin Gap Chambers. This paper describes the construction and cosmic-ray testing of small-strip Thin Gap Chambers in Shandong University.

physics.ins-det↗

arXiv2018

MWPC prototyping and performance test for the STAR inner TPC upgrade

A new prototype of STAR inner Time Projection Chamber (iTPC) MWPC sector has been fabricated and tested in an X-ray test system. The wire chamber built at Shandong University has a wire tension precision better than 6$\%$ and wire pitch precision better than 10 $μ$m. The gas gain uniformity and energy resolution are measured to be better than 1$\%$ (RMS) and 20$\%$ (FWHM), respectively, using an $^{55}$Fe X-ray source. The iTPC upgrade project is to replace all 24 STAR TPC inner sectors as a crucial detector upgrade for the RHIC beam energy scan phase II program. The test results show that the constructed iTPC prototype meets all project requirements.

physics.ins-det↗

arXiv2017

Design and implementation of wire tension measurement system for MWPCs used in the STAR iTPC upgrade

The STAR experiment at RHIC is planning to upgrade the Time Projection Chamber which lies at the heart of the detector. We have designed an instrument to measure the tension of the wires in the multi-wire proportional chambers (MWPCs) which will be used in the TPC upgrade. The wire tension measurement system causes the wires to vibrate and then it measures the fundamental frequency of the oscillation via a laser based optical platform. The platform can scan the entire wire plane, automatically, in a single run and obtain the wire tension on each wire with high precision. In this paper, the details about the measurement method and the system setup will be described. In addition, the test results for a prototype MWPC to be used in the STAR-iTPC upgrade will be presented.

physics.ins-det↗

arXiv2015

Performance of a Full-Size Small-Strip Thin Gap Chamber Prototype for the ATLAS New Small Wheel Muon Upgrade

The instantaneous luminosity of the Large Hadron Collider at CERN will be increased up to a factor of five with respect to the present design value by undergoing an extensive upgrade program over the coming decade. The most important upgrade project for the ATLAS Muon System is the replacement of the present first station in the forward regions with the so-called New Small Wheels (NSWs). The NSWs will be installed during the LHC long shutdown in 2018/19. Small-Strip Thin Gap Chamber (sTGC) detectors are designed to provide fast trigger and high precision muon tracking under the high luminosity LHC conditions. To validate the design, a full-size prototype sTGC detector of approximately 1.2 $\times$ $1.0\, \mathrm{m}^2$ consisting of four gaps has been constructed. Each gap provides pad, strip and wire readouts. The sTGC intrinsic spatial resolution has been measured in a $32\, \mathrm{GeV}$ pion beam test at Fermilab. At perpendicular incidence angle, single gap position resolutions of about $50\,\mathrm{μm}$ have been obtained, uniform along the sTGC strip and perpendicular wire directions, well within design requirements. Pad readout measurements have been performed in a $130\, \mathrm{GeV}$ muon beam test at CERN. The transition region between readout pads has been found to be $4\,\mathrm{mm}$, and the pads have been found to be fully efficient.

physics.ins-det↗