arXiv subjects

Jingzhe Tang

Publications and source records attributed to Jingzhe Tang.

4 recordsLinked to original sources

arXiv2026

Simulation Based Characterization of Deconvolution-Based PMT Waveform Reconstruction Under Large Charge Dynamic Range and Varying Scintillation Time Profiles

Photomultiplier tubes (PMTs) are widely used as photon sensors for neutrino and dark matter detection. Accurate charge and time information extracted from PMT waveforms is crucial for event reconstruction. An algorithm based on deconvolution technology was proposed and applied to the reconstruction of PMT waveforms. This study further investigated the reliability of the deconvolution algorithm when handling a large charge dynamic range (0-200 photoelectrons), varying scintillation time profiles, and muon-induced large signals. Monte Carlo data confirmed that the deconvolution algorithm exhibits relatively stable reconstruction performance: under the simulation conditions described in this paper (including a noise level of 0.1 PE, single photoelectron charge resolution of 30%, 1 GHz sampling rate, 1000 ns window, three undershoot configurations, and eight scintillation time profiles), the residual non-linearity of charge reconstruction is controlled to approximately 1% over the range of 0 to 200 photoelectrons, and the algorithm is capable of handling muon-induced large signals. The reconstruction performance depends on adequate baseline recovery; a waveform window that is too short relative to the undershoot tail leads to degraded reconstruction quality, which can be mitigated by extending the sampling window.

physics.ins-det↗

arXiv2025

Robust LLM Training Infrastructure at ByteDance

The training scale of large language models (LLMs) has reached tens of thousands of GPUs and is still continuously expanding, enabling faster learning of larger models. Accompanying the expansion of the resource scale is the prevalence of failures (CUDA error, NaN values, job hang, etc.), which poses significant challenges to training stability. Any large-scale LLM training infrastructure should strive for minimal training interruption, efficient fault diagnosis, and effective failure tolerance to enable highly efficient continuous training. This paper presents ByteRobust, a large-scale GPU infrastructure management system tailored for robust and stable training of LLMs. It exploits the uniqueness of LLM training process and gives top priorities to detecting and recovering failures in a routine manner. Leveraging parallelisms and characteristics of LLM training, ByteRobust enables high-capacity fault tolerance, prompt fault demarcation, and localization with an effective data-driven approach, comprehensively ensuring continuous and efficient training of LLM tasks. ByteRobust is deployed on a production GPU platform and achieves 97% ETTR for a three-month training job on 9,600 GPUs.

cs.LG↗

arXiv2025

The High Voltage Splitter board for the JUNO SPMT system

The Jiangmen Underground Neutrino Observatory (JUNO) in southern China is designed to study neutrinos from nuclear reactors and natural sources to address fundamental questions in neutrino physics. Achieving its goals requires continuous operation over a 20-year period. The small photomultiplier tube (small PMT or SPMT) system is a subsystem within the experiment composed of 25600 3-inch PMTs and their associated readout electronics. The High Voltage Splitter (HVS) is the first board on the readout chain of the SPMT system and services the PMTs by providing high voltage for biasing and by decoupling the generated physics signal from the high-voltage bias for readout, which is then fed to the front-end board. The necessity to handle high voltage, manage a large channel count, and operate stably for 20 years imposes significant constraints on the physical design of the HVS. This paper serves as a comprehensive documentation of the HVS board: its role in the SPMT readout system, the challenges in its design, performance and reliability metrics, and the methods employed for production and quality control.

physics.ins-det↗

arXiv2025

Investigation and optimization of the deconvolution method for PMT waveform reconstruction

Photomultiplier tubes (PMTs) are extensively employed as photosensors in neutrino and dark matter detection. The precise charge and timing information extracted from the PMT waveform plays a crucial role in energy and vertex reconstruction. In this study, we investigate the deconvolution algorithm utilized for PMT waveform reconstruction, while enhancing the timing separation ability for pile-up hits by redesigning filters based on the time-frequency uncertainty principle. This filter design sacrifices signal-to-noise ratio (SNR) to achieve narrower pulse widths. Furthermore, we optimize the selection of signal pulses in the case of low SNR based on Short-Time Fourier Transform (STFT). Monte Carlo data confirms that our optimization yields enhanced reconstruction performance: improving timing separation ability for pile-up hits from $7\sim10$~ns to $3\sim5$~ns, while controlling the residual nonlinearity of charge reconstruction to about 1\% in the range of 0 to 20 photoelectrons.

physics.ins-det↗