arXiv subjects

Shiqi Kuang

Publications and source records attributed to Shiqi Kuang.

7 recordsLinked to original sources

arXiv2026

On-Detector Machine Learning for Beam-Induced Background Rejection at a 10 TeV Muon Collider

A 10 TeV Muon Collider is a compelling candidate for a future energy-frontier facility, offering unprecedented opportunities to explore the fundamental laws of particle physics. Muon decays in the collider ring produce intense beam-induced background (BIB) that can overwhelm detector occupancy and exceed readout bandwidth constraints. We investigate the potential of on-detector Machine Learning for BIB rejection in the vertex detector, exploiting pixel cluster shapes to distinguish background from collision products. We study three classes of lightweight neural-network architectures, and evaluate their implementation feasibility using high-level synthesis. Selected architectures achieve 88 to 90% data reduction at 99% signal efficiency, while requiring hardware resources compatible with potential ASIC implementation. These results demonstrate the potential of performing substantial BIB rejection directly in the pixel readout, providing a strategy for meeting the tracker readout requirements at a future Muon Collider.

hep-ex↗

arXiv2026

REAgent: Requirement-Driven LLM Agents for Software Issue Resolution

Issue resolution aims to automatically generate patches from given issue descriptions and has attracted significant attention with the rapid advancement of large language models (LLMs). However, due to the complexity of software issues and codebases, LLM-generated patches often fail to resolve corresponding issues. Although various advanced techniques have been proposed with carefully designed tools and workflows, they typically treat issue descriptions as direct inputs and largely overlook their quality (e.g., missing critical context or containing ambiguous information), which hinders LLMs from accurate understanding and resolution. To address this limitation, we draw on principles from software requirements engineering and propose REAgent, a requirement-driven LLM agent framework that introduces issue-oriented requirements as structured task specifications to better guide patch generation. Specifically, REAgent automatically constructs structured and information-rich issue-oriented requirements, identifies low-quality requirements, and iteratively refines them to improve patch correctness. We conduct comprehensive experiments on three widely used benchmarks using two advanced LLMs, comparing against five representative or state-of-the-art baselines. The results demonstrate that REAgent consistently outperforms all baselines, achieving an average improvement of 17.40% in terms of the number of successfully-resolved issues (% Resolved).

cs.SE↗

arXiv2026

On-chip probabilistic inference for charged-particle tracking at the sensor edge

Modern scientific instruments operate under increasingly extreme constraints on bandwidth, latency, and power. Inference at the sensor edge determines experimental data collection efficiency by deciding which information to save for further analysis. Particle tracking detectors at the Large Hadron Collider exemplify this challenge: pixelated silicon sensors generate rich spatiotemporal ionization patterns, yet most of this information is discarded due to data-rate limitations. Concurrently, advancements in co-design tools provide rapid turn-around for incorporating machine learning into application-specific integrated circuits, motivating designs for particle detectors with new integrated technologies. We demonstrate that neural networks embedded in the front-end electronics can infer charged-particle kinematic parameters from a single silicon layer. We regress hit positions and incident angles with calibrated uncertainties, while satisfying stringent constraints on numerical precision, latency, and silicon area. Our results establish a path toward probabilistic inference directly at the edge, opening new opportunities for intelligent sensing in high-rate scientific instruments.

physics.ins-det↗

arXiv2025

On the Effectiveness of Training Data Optimization for LLM-based Code Generation: An Empirical Study

Large language models (LLMs) have achieved remarkable progress in code generation, largely driven by the availability of high-quality code datasets for effective training. To further improve data quality, numerous training data optimization techniques have been proposed; however, their overall effectiveness has not been systematically evaluated. To bridge this gap, we conduct the first large-scale empirical study, examining five widely-used training data optimization techniques and their pairwise combinations for LLM-based code generation across three benchmarks and four LLMs. Our results show that data synthesis is the most effective technique for improving functional correctness and reducing code smells, although it performs relatively worse on code maintainability compared to data refactoring, cleaning, and selection. Regarding combinations, we find that most combinations do not further improve functional correctness but can effectively enhance code quality (code smells and maintainability). Among all combinations, data synthesis combined with data refactoring achieves the strongest overall performance. Furthermore, our fine-grained analysis reinforces these findings and provides deeper insights into how individual techniques and their combinations influence code generation effectiveness. Overall, this work represents a first step toward a systematic understanding of training data optimization and combination strategies, offering practical guidance for future research and deployment in LLM-based code generation.

cs.SE↗

arXiv2025

Sensor Co-design for $\textit{smartpixels}$

Pixel tracking detectors at upcoming collider experiments will see unprecedented charged-particle densities. Real-time data reduction on the detector will enable higher granularity and faster readout, possibly enabling the use of the pixel detector in the first level of the trigger for a hadron collider. This data reduction can be accomplished with a neural network (NN) in the readout chip bonded with the sensor that recognizes and rejects tracks with low transverse momentum (p$_T$) based on the geometrical shape of the charge deposition (``cluster''). To design a viable detector for deployment at an experiment, the dependence of the NN as a function of the sensor geometry, external magnetic field, and irradiation must be understood. In this paper, we present first studies of the efficiency and data reduction for planar pixel sensors exploring these parameters. A smaller sensor pitch in the bending direction improves the p$_T$ discrimination, but a larger pitch can be partially compensated with detector depth. An external magnetic field parallel to the sensor plane induces Lorentz drift of the electron-hole pairs produced by the charged particle, broadening the cluster and improving the network performance. The absence of the external field diminishes the background rejection compared to the baseline by $\mathcal{O}$(10%). Any accumulated radiation damage also changes the cluster shape, reducing the signal efficiency compared to the baseline by $\sim$ 30 - 60%, but nearly all of the performance can be recovered through retraining of the network and updating the weights. Finally, the impact of noise was investigated, and retraining the network on noise-injected datasets was found to maintain performance within 6% of the baseline network trained and evaluated on noiseless data.

physics.ins-det↗

arXiv2025

SRPO: A Cross-Domain Implementation of Large-Scale Reinforcement Learning on LLM

Recent advances of reasoning models, exemplified by OpenAI's o1 and DeepSeek's R1, highlight the significant potential of Reinforcement Learning (RL) to enhance the reasoning capabilities of Large Language Models (LLMs). However, replicating these advancements across diverse domains remains challenging due to limited methodological transparency. In this work, we present two-Staged history-Resampling Policy Optimization (SRPO), which surpasses the performance of DeepSeek-R1-Zero-32B on the AIME24 and LiveCodeBench benchmarks. SRPO achieves this using the same base model as DeepSeek (i.e. Qwen2.5-32B), using only about 1/10 of the training steps required by DeepSeek-R1-Zero-32B, demonstrating superior efficiency. Building upon Group Relative Policy Optimization (GRPO), we introduce two key methodological innovations: (1) a two-stage cross-domain training paradigm designed to balance the development of mathematical reasoning and coding proficiency, and (2) History Resampling (HR), a technique to address ineffective samples. Our comprehensive experiments validate the effectiveness of our approach, offering valuable insights into scaling LLM reasoning capabilities across diverse tasks.

cs.LG↗

arXiv2024

On-demand Quick Metasurface Design with Neighborhood Attention Transformer

Metasurfaces are reshaping traditional optical paradigms and are increasingly required in complex applications that demand substantial computational resources to numerically solve Maxwell's equations-particularly for large-scale systems, inhomogeneous media, and densely packed metadevices. Conventional forward design using electromagnetic solvers is based on specific approximations, which may not effectively address complex problems. In contrast, existing inverse design methods are a stepwise process that is often time-consuming and involves repetitive computations. Here, we present an inverse design approach utilizing a surrogate Neighborhood Attention Transformer, MetaE-former, to predict the performance of metasurfaces with ultrafast speed and high accuracy. This method achieves global solutions for hundreds of nanostructures simultaneously, providing up to a 250,000-fold speedup compared with solving for individual meta-atoms based on the FDTD method. As examples, we demonstrate a binarized high-numerical-aperture (about 1.31) metalens and several optimized structured-light meta-generators. Our method significantly improves the beam shaping adaptability with metasurfaces and paves the way for fast designing of large-scale metadevices for shaping extreme light fields with high accuracy.

physics.optics↗