Searcharxiv⌕ Search

arXiv subjects

Meiling Chen

Publications and source records attributed to Meiling Chen.

8 recordsLinked to original sources

Reducing Batch Makespan for LLM Agent Workloads through Inference-Environment Coordination

Efficient batch execution of agent tasks is critical to LLM training and development workflows, including benchmark evaluation, synthetic-data generation, and RL rollouts. Since these tasks alternate between LLM inference and environment interaction, we present coordinated optimizations to reduce batch makespan, the time required to complete all tasks in a batch. For hybrid LLMs, a workload-based resource model guides GPU-memory allocation between key-value (KV) cache and recurrent state and provides a reference sample concurrency to exploit parallelism while limiting recomputation. Trajectory-aware routing and scheduling promote cache reuse while balancing worker load. Under constrained environment capacity, we decouple sample admission from environment execution and introduce Chain-Advancing Action Scheduling (CAAS), which prioritizes long chains using observed progress to shorten the batch tail. Controlled replay combines actual inference with recorded environment responses and durations for ablations of individual optimizations. Live evaluations with Ling-3.0-Flash and Ling-3.1-Flash on the BFCL, SWE-bench Multilingual, WideSearch, and MCP-Atlas benchmarks demonstrate makespan reductions of 10.4% to 65.0%, while evaluation scores remain aligned. Furthermore, these methods have been successfully applied to training and evaluation workflows for the Ling-3.x model series, accelerating the model-development cycle.

cs.DC↗

Development and Performance Study of a Capillary Liquid Scintillator Neutron Detector

Capillary liquid scintillator detectors are promising for high-resolution neutron imaging, yet experimental data on their light spread mechanism and spatial performance remain limited. Here, we report a neutron detector based on a hexagonal capillary array filled with EJ-309 liquid scintillator, with an inner diameter of about 50 um and a camera readout of 9 um pixels. Laser experiments show that the FWHM of the full light spot decreases from 260 um to 90 um with a metal light absorber, confirming effective suppression of lateral light spread. Using an AmBe neutron source, an effective field of view with a 5-sigma threshold was established from background frames. For single-capillary events, the pulse height spectrum follows a Landau distribution with a most probable value of 0.133 +/- 0.001 (stat.), and the intrinsic detection efficiency is 10.07% +/- 1.26% (stat.) +/- 1.43% (syst.), corresponding to about 13.55% when normalized to the active liquid scintillator area. The point spread function core yields a radial FWHM of 12.8 um and a centroid positioning precision of approximately 5.5 um (1 sigma), while the intrinsic position resolution is limited by the capillary pitch to 54 um. Linearity is good for 1 to 2 capillaries, with deviation appearing for 2 to 3 capillaries due to additional capture of spread light. These results provide experimental basis and physical understanding for imaging applications of capillary liquid scintillator neutron detectors.

physics.ins-det↗

Loop Engineering: Version-Bound, Evidence-Gated Coordination for Production Coding Agents

Coding agents can edit repositories and execute tests, yet production software work extends far beyond code generation. It requires durable task context, scoped authority, version-aware approval, source-independent verification, interruption recovery, and accountable handoffs between humans and automated workers. We present Loop Engineering, a method for operating multiple asynchronous agent loops over shared, auditable repository state without relying on private conversational handoffs. Each loop is specified by six elements: a trigger, a context projection, an authority boundary, a verifier, a stop condition, and a write-back rule. Versioned procedural contracts, called Skills, encode reusable operating policy; a Harness bounds the available action surface and collects evidence; human gates authorize high-consequence transitions against exact artifact versions. We report a design-and-experience study from MewDesign, a revenue-generating commercial AI design product. Five specialized loops cover planning, scoped implementation, review, test release, and reconciliation. A cross-sectional audit contains 50 durable marker records across 26 GitHub work items. Five pull requests reached a version-bound test lock: three proceeded to test releases with runtime evidence from sources distinct from the deployment request, and two stopped before merge. A detailed drift episode shows one observed fail-closed path: a new commit invalidated prior approval and evidence, and the workflow refused to reuse stale continuous-integration results. The study is not a controlled benchmark and does not establish causal productivity or general safety. It provides early evidence of feasibility, traceability, and bounded progression in one production setting.

cs.SE↗

ITKIT: Feasible CT Image Analysis based on SimpleITK and MMEngine

CT images are widely used in clinical diagnosis and treatment, and their data have formed a de facto standard - DICOM. It is clear and easy to use, and can be efficiently utilized by data-driven analysis methods such as deep learning. In the past decade, many program frameworks for medical image analysis have emerged in the open-source community. ITKIT analyzed the characteristics of these frameworks and hopes to provide a better choice in terms of ease of use and configurability. ITKIT offers a complete pipeline from DICOM to 3D segmentation inference. Its basic practice only includes some essential steps, enabling users with relatively weak computing capabilities to quickly get started using the CLI according to the documentation. For advanced users, the OneDL-MMEngine framework provides a flexible model configuration and deployment entry. This paper conducted 12 typical experiments to verify that ITKIT can meet the needs of most basic scenarios.

cs.SE↗

Spatial-Aware Self-Supervision for Medical 3D Imaging with Multi-Granularity Observable Tasks

The application of self-supervised techniques has become increasingly prevalent within medical visualization tasks, primarily due to its capacity to mitigate the data scarcity prevalent in the healthcare sector. The majority of current works are influenced by designs originating in the generic 2D visual domain, which lack the intuitive demonstration of the model's learning process regarding 3D spatial knowledge. Consequently, these methods often fall short in terms of medical interpretability. We propose a method consisting of three sub-tasks to capture the spatially relevant semantics in medical 3D imaging. Their design adheres to observable principles to ensure interpretability, and minimize the performance loss caused thereby as much as possible. By leveraging the enhanced semantic depth offered by the extra dimension in 3D imaging, this approach incorporates multi-granularity spatial relationship modeling to maintain training stability. Experimental findings suggest that our approach is capable of delivering performance that is on par with current methodologies, while facilitating an intuitive understanding of the self-supervised learning process.

cs.CV↗

Intuitive Axial Augmentation Using Polar-Sine-Based Piecewise Distortion for Medical Slice-Wise Segmentation

Most data-driven models for medical image analysis rely on universal augmentations to improve accuracy. Experimental evidence has confirmed their effectiveness, but the unclear mechanism underlying them poses a barrier to the widespread acceptance and trust in such methods within the medical community. We revisit and acknowledge the unique characteristics of medical images apart from traditional digital images, and consequently, proposed a medical-specific augmentation algorithm that is more elastic and aligns well with radiology scan procedure. The method performs piecewise affine with sinusoidal distorted ray according to radius on polar coordinates, thus simulating uncertain postures of human lying flat on the scanning table. Our method could generate human visceral distribution without affecting the fundamental relative position on axial plane. Two non-adaptive algorithms, namely Meta-based Scan Table Removal and Similarity-Guided Parameter Search, are introduced to bolster robustness of our augmentation method. In contrast to other methodologies, our method is highlighted for its intuitive design and ease of understanding for medical professionals, thereby enhancing its applicability in clinical scenarios. Experiments show our method improves accuracy with two modality across multiple famous segmentation frameworks without requiring more data samples. Our preview code is available in: https://github.com/MGAMZ/PSBPD.

cs.CV↗

Interpretable Auto Window Setting for Deep-Learning-Based CT Analysis

Whether during the early days of popularization or in the present, the window setting in Computed Tomography (CT) has always been an indispensable part of the CT analysis process. Although research has investigated the capabilities of CT multi-window fusion in enhancing neural networks, there remains a paucity of domain-invariant, intuitively interpretable methodologies for Auto Window Setting. In this work, we propose an plug-and-play module originate from Tanh activation function, which is compatible with mainstream deep learning architectures. Starting from the physical principles of CT, we adhere to the principle of interpretability to ensure the module's reliability for medical implementations. The domain-invariant design facilitates observation of the preference decisions rendered by the adaptive mechanism from a clinically intuitive perspective. This enables the proposed method to be understood not only by experts in neural networks but also garners higher trust from clinicians. We confirm the effectiveness of the proposed method in multiple open-source datasets, yielding 10%~200% Dice improvements on hard segment targets.

eess.IV↗

Extension of a Pattern Recognition Validation Approach for Noisy Boson Sampling

Boson sampling is one of the main quantum computation models to demonstrate the quantum computational advantage. However, this aim may be hard to realize considering two main kinds of noises, which are photon distinguishability and photon loss. Inspired by the Bayesian validation extended to evaluate whether distinguishability is too high to demonstrate this advantage, the pattern recognition validation is extended for boson sampling, considering both distinguishability and loss. Based on clusters constructed with the K means++ method, where parameters are carefully adjusted to optimize the extended validation performances, the distribution of characteristic values is nearly monotonically changed with indistinguishability, especially when photons are close to be indistinguishable. However, this regulation may be suppressed by photon loss. The intrinsic data structure of output events is analyzed through calculating probability distributions and mean 2-norm distances of the sorted outputs. An approximation algorithm is also used to show the data structure changes with noises.

quant-ph↗