arXiv subjects

Zhuo Zuo

Publications and source records attributed to Zhuo Zuo.

6 recordsLinked to original sources

arXiv2026

Enhancing Numerical Prediction in LLMs via Smooth MMD Alignment

Despite their strong general capabilities, large language models (LLMs) often remain unreliable when outputs must be numerically precise. A key reason is the training objective: standard cross-entropy treats numeric tokens as unstructured categories and ignores the metric structure of their values. We address this mismatch with Smooth Maximum Mean Discrepancy (SMMD), which builds on the classic MMD by incorporating value-distance kernels over numeric tokens and graph-based smoothness. With this kernel defined over a numeric sub-vocabulary, SMMD aligns the predicted numeric distribution to the target via kernel matching and smooths the prediction-target residual over the induced kernel graph to encourage local consistency. We evaluate SMMD on four numeric-target tasks: mathematical reasoning, arithmetic calculation, clock-time recognition, and chart question answering, across multiple open-weight LLM and VLM backbones. SMMD consistently improves accuracy over both cross-entropy and recent numeric-target losses; analyses show complementary effects between MMD and smoothness and underscore the importance of distance-based kernel design. Code is available at https://github.com/Zuozhuo/smmd-loss.

cs.CL↗

arXiv2025

Pulse Shape Discrimination Algorithms: Survey and Benchmark

This review presents a comprehensive survey and benchmark of pulse shape discrimination (PSD) algorithms for radiation detection, classifying nearly sixty methods into statistical (time-domain, frequency-domain, neural network-based) and prior-knowledge (machine learning, deep learning) paradigms. We implement and evaluate all algorithms on two standardized datasets: an unlabeled set from a 241Am-9Be source and a time-of-flight labeled set from a 238Pu-9Be source, using metrics including Figure of Merit (FOM), F1-score, ROC-AUC, and inter-method correlations. Our analysis reveals that deep learning models, particularly Multi-Layer Perceptrons (MLPs) and hybrid approaches combining statistical features with neural regression, often outperform traditional methods. We discuss architectural suitabilities, the limitations of FOM, alternative evaluation metrics, and performance across energy thresholds. Accompanying this work, we release an open-source toolbox in Python and MATLAB, along with the datasets, to promote reproducibility and advance PSD research.

cs.LG↗

arXiv2024

Dataset for neutron and gamma-ray pulse shape discrimination

The publicly accessible dataset includes neutron and gamma-ray pulse signals for conducting pulse shape discrimination experiments. Several traditional and recently proposed pulse shape discrimination algorithms are utilized to evaluate the performance of pulse shape discrimination under raw pulse signals and noise-enhanced datasets. These algorithms comprise zero-crossing (ZC), charge comparison (CC), falling edge percentage slope (FEPS), frequency gradient analysis (FGA), pulse-coupled neural network (PCNN), ladder gradient (LG), and het-erogeneous quasi-continuous spiking cortical model (HQC-SCM). In addition to the pulse signals, this dataset includes the source code for all the aforementioned pulse shape discrimination methods. Moreover, the dataset provides the source code for schematic pulse shape discrimination performance evaluation and anti-noise performance evaluation. This feature enables researchers to evaluate the performance of these methods using standard procedures and assess their anti-noise ability under various noise conditions. In conclusion, this dataset offers a comprehensive set of resources for conducting pulse shape discrimination experiments and evaluating the performance of various pulse shape discrimination methods under different noise scenarios.

eess.SP↗

arXiv2024

Random-coupled Neural Network

Improving the efficiency of current neural networks and modeling them in biological neural systems have become popular research directions in recent years. Pulse-coupled neural network (PCNN) is a well applicated model for imitating the computation characteristics of the human brain in computer vision and neural network fields. However, differences between the PCNN and biological neural systems remain: limited neural connection, high computational cost, and lack of stochastic property. In this study, random-coupled neural network (RCNN) is proposed. It overcomes these difficulties in PCNN's neuromorphic computing via a random inactivation process. This process randomly closes some neural connections in the RCNN model, realized by the random inactivation weight matrix of link input. This releases the computational burden of PCNN, making it affordable to achieve vast neural connections. Furthermore, the image and video processing mechanisms of RCNN are researched. It encodes constant stimuli as periodic spike trains and periodic stimuli as chaotic spike trains, the same as biological neural information encoding characteristics. Finally, the RCNN is applicated to image segmentation, fusion, and pulse shape discrimination subtasks. It is demonstrated to be robust, efficient, and highly anti-noised, with outstanding performance in all applications mentioned above.

cs.CV↗

arXiv2023

Pulse shape discrimination based on the Tempotron: a powerful classifier on GPU

This study utilized the Tempotron, a robust classifier based on a third-generation neural network model, for pulse shape discrimination. By eliminating the need for manual feature extraction, the Tempotron model can process pulse signals directly, generating discrimination results based on prior knowledge. The study performed experiments using GPU acceleration, resulting in over 500 times faster compared to the CPU-based model, and investigated the impact of noise augmentation on the Tempotron performance. Experimental results substantiated that Tempotron serves as a formidable classifier, adept at accomplishing high discrimination accuracy on both AmBe and time-of-flight PuBe datasets. Furthermore, analyzing the neural activity of Tempotron during training shed light on its learning characteristics and aided in selecting its hyperparameters. Moreover, the study addressed the constraints and potential avenues for future development in utilizing the Tempotron for pulse shape discrimination. The dataset used in this study and the GPU-based Tempotron are publicly available on GitHub at https://github.com/HaoranLiu507/TempotronGPU.

eess.SP↗

arXiv2016

Toward a fractal spectrum approach for neutron and gamma pulse shape discrimination

There is a key research issue to accurately select out neutron signals and discriminate gamma signals from a mixed radiation field in the neutron detection. This paper proposes a fractal spectrum discrimination approach by means of different spectrum characteristics of neutron and gamma. Figure of merit and average discriminant error ratio are adopted together to evaluate the discriminant effects. Different neutron and gamma signals with various noises and pulse pile-ups are simulated according to real data in the literature. The proposed approach is compared with the digital charge integration and pulse gradient methods. It is found that the fractal approach exhibits the best discriminant performance among three methods. The fractal spectrum approach is not sensitive to the high frequency noises and pulse pile-ups. It means that the proposed approach takes the advantages of anti-noises and high discriminant ability, and can be used to better discriminate neutron and gamma in neutron detection.

physics.ins-det↗