arXiv subjects

Sitian Qian

Publications and source records attributed to Sitian Qian.

At least 19 recordsLinked to original sources

arXiv2026

SE-GoS: Self-Evolving Graph-of-Skills for Skill Library at Scale

Modern LLM agents increasingly rely on reusable skills, yet as skill libraries scale to thousands of entries, effective retrieval becomes a bottleneck. Graph-of-Skills (GoS) addresses this challenge by exploiting dependency-aware graph structure for scalable skill retrieval, while SkillDAG further demonstrates that skill graphs can accumulate execution-backed structure online. However, these approaches leave open whether historical execution traces can be systematically distilled into a better retrieval graph that generalizes to unseen tasks. We present Self-Evolving Graph-of-Skills (SE-GoS), a training-free framework that evolves an existing GoS graph from execution traces while preserving the original retrieval pipeline. SE-GoS performs three complementary updates: topology evolution that discovers and prunes skill relationships from execution evidence, edge-weight evolution that reinforces retrieval-relevant relationships based on historical effectiveness, and description evolution that optimizes retrieval-facing skill descriptions using execution feedback. Across three LLMs on SkillsBench, SE-GoS consistently improves task reward while reducing input tokens relative to full skill loading, with gains varying across model families. In a representative setting, one evolution round improves reward from 52.4\% to 59.4\% while reducing input tokens by approximately one-third relative to full skill loading, and the resulting graph transfers to a disjoint held-out split with a 5.4-point improvement over the static GoS baseline. These results show that skill graphs can be improved from execution experience without model training, changes to the retrieval algorithm, or modifications to skill content, turning a static retrieval graph into an evolving retrieval infrastructure.

cs.AI↗

arXiv2026

CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters

High-precision calorimeter simulation at current and future colliders imposes rapidly growing computational demands, motivating the development of machine-learning surrogates for traditional Monte Carlo tools such as Geant4. Flow matching and diffusion-based generative models have become leading approaches for high-dimensional fast simulation because of their sample quality, but typically require ${\cal O}(100)$ function evaluations at inference and often rely on auxiliary networks to constrain global observables, compromising streamlined end-to-end generation. We introduce a unified framework that improves the balance between speed, shower quality, and physics fidelity. The method combines: (i) an average velocity field integrator that enables sampling in one or a few evaluations; (ii) a learned generative prior in shower space, constructed from data rather than random noise; and (iii) physics-guided loss terms that impose inductive biases on key observables during training. These elements are training time regularizers, preserving end-to-end inference with no additional cost. With only one or a few evaluation steps, the model achieves shower quality competitive with state-of-the-art flow and diffusion approaches, tested on several public high granularity calorimeter datasets. The results demonstrate inter-layer shower structure consistent with the underlying physics, providing a strong candidate for future fast simulation workflows.

hep-ex↗

arXiv2026

Distilling the knowledge with quantum neural networks

Quantum Neural Networks (QNNs) are a promising class of quantum machine learning models with potential quantum advantages when implemented on scalable, error-corrected quantum computers. However, as system sizes increase, deploying QNNs becomes challenging. Similar to their classical counterparts, a key obstacle to their practical applications is that large-scale QNNs may not be easily deployed on smaller systems that have limited resources. Here, we tackle this challenge by compressing QNNs via knowledge distillation. We demonstrate how well-trained QNNs on large systems can be distilled into smaller architectures with similar configurations. We numerically show that knowledge distillation helps reduce the training cost of QNNs in terms of the number of qubits and circuit depth. Additionally, we find that a self-knowledge-distillation approach can accelerate training convergence. We believe our results offer new strategies for the efficient compression and practical deployment of QNNs.

quant-ph↗

arXiv2025

Quantum Entanglement between gauge boson pairs at a Muon Collider

Quantum entanglement is one of significant physics phenomena that can be examined at a particle collider. A muon collider can provide a stage on which we can study substantial physics phenomenon, starting from the precision measurements of the Standard Model and beyond to the undiscovered area of physics. In this work, we present a through study of quantum entanglement in $\mu^+\mu^-\to ZZ$ events at a future muon collider. By fixing the spin density matrix, observables quantifying entanglement between $Z$ boson pairs can be measured. After systematic Monte-Carlo simulation and background analysis, we measure the value of entanglement variables and perform hypothesis testing against the non-entangled hypothesis, finally observing the entanglement of the $ZZ$ system up to $2$ significance level.

hep-ph↗

arXiv2025

Application of Structured State Space Models to High energy physics with locality-sensitive hashing

Modern high-energy physics (HEP) experiments are increasingly challenged by the vast size and complexity of their datasets, particularly regarding large-scale point cloud processing and long sequences. In this study, to address these challenges, we explore the application of structured state space models (SSMs), proposing one of the first trials to integrate local-sensitive hashing into either a hybrid or pure Mamba Model. Our results demonstrate that pure SSMs could serve as powerful backbones for HEP problems involving tasks for long sequence data with local inductive bias. By integrating locality-sensitive hashing into Mamba blocks, we achieve significant improvements over traditional backbones in key HEP tasks, surpassing them in inference speed and physics metrics while reducing computational overhead. In key tests, our approach demonstrated promising results, presenting a viable alternative to traditional transformer backbones by significantly reducing FLOPS while maintaining robust performance.

cs.LG↗

arXiv2024

Testing Bell inequalities and probing quantum entanglement at a muon collider

A muon collider represents a promising candidate for the next generation of particle physics experiments after the expected end of LHC operations in the early 2040s. Rare or hard-to-detect processes at the LHC, such as the production of multiple gauge bosons, become accessible at a TeV muon collider. We present here the prospects of detecting quantum entanglement and the violation of Bell inequalities in H to ZZ to 4l events at a potential future muon collider. We show that the spin density matrix of the Z boson pairs can be reconstructed using the kinematics of the charged leptons from the Z boson decays. Once the density matrix is determined, it is straightforward to obtain the expectation values of various Bell operators and test the quantum entanglement between the Z boson pair. Through a detailed study based on Monte-Carlo simulation, we show that the generalized CGLMP inequality can be maximally violated, and testing Bell inequalities could be established with high significance.

hep-ph↗

arXiv2024

BUFF: Boosted Decision Tree based Ultra-Fast Flow matching

Tabular data stands out as one of the most frequently encountered types in high energy physics. Unlike commonly homogeneous data such as pixelated images, simulating high-dimensional tabular data and accurately capturing their correlations are often quite challenging, even with the most advanced architectures. Based on the findings that tree-based models surpass the performance of deep learning models for tasks specific to tabular data, we adopt the very recent generative modeling class named conditional flow matching and employ different techniques to integrate the usage of Gradient Boosted Trees. The performances are evaluated for various tasks on different analysis level with several public datasets. We demonstrate the training and inference time of most high-level simulation tasks can achieve speedup by orders of magnitude. The application can be extended to low-level feature simulation and conditioned generations with competitive performance.

physics.ins-det↗

arXiv2024

Choose Your Diffusion: Efficient and flexible ways to accelerate the diffusion model in fast high energy physics simulation

The diffusion model has demonstrated promising results in image generation, recently becoming mainstream and representing a notable advancement for many generative modeling tasks. Prior applications of the diffusion model for both fast event and detector simulation in high energy physics have shown exceptional performance, providing a viable solution to generate sufficient statistics within a constrained computational budget in preparation for the High Luminosity LHC. However, many of these applications suffer from slow generation with large sampling steps and face challenges in finding the optimal balance between sample quality and speed. The study focuses on the latest benchmark developments in efficient ODE/SDE-based samplers, schedulers, and fast convergence training techniques. We test on the public CaloChallenge and JetNet datasets with the designs implemented on the existing architecture, the performance of the generated classes surpass previous models, achieving significant speedup via various evaluation metrics.

physics.ins-det↗

arXiv2023

Searching for Majorana Neutrinos at a Same-Sign Muon Collider

Majorana properties of neutrinos have long been a focus in the pursuit of possible new physics beyond the standard model, which has motivated lots of dedicated theoretical and experimental studies. A future same-sign muon collider is an ideal platform to search for Majorana neutrinos through the Lepton Number Violation process. Specifically, this t-channel kind of process is less kinematically suppressed and has a good advantage in probing Majorana neutrinos at high mass regions up to 10 TeV. In this paper, we perform a detailed fast Monte Carlo simulation study through examining three different final states: 1) pure-leptonic state with electrons or muons, 2) semi-leptonic state, and 3) pure-hadronic state in the resolved or merged categories. Furthermore, we perform a full simulation study on the pure-leptonic final state to validate our fast simulation results.

hep-ph↗

arXiv2023

Low Energy Neutrino and Mass Dark Matter Detection Using Freely Falling Atoms

We propose a new method to detect low-energy neutrinos and low-mass dark matter at or below the MeV scale, through their coherent scatterings from freely falling heavy atoms and the resulting kinematic shifts. We start with a simple calculation for illustration: for $10^7$ heavy atoms of a mass number around 100 with a small recoil energy of 1 meV, the corresponding velocities can reach $0.01, {\rm m/s}$ and produce significant kinematic shifts that can be detected. We then show that the proposed device should be able to probe vast low-energy regions of neutrinos from meV to MeV and can surpass previous limits on sub-MeV dark matter by several orders of magnitude. Such a proposal can be useful to (1) detect sub-MeV-scale dark matter: with $10^2$ atom guns shooting downwards, for example, CsI or lead clusters consisting of $10^{7}$ atoms with a frequency around $10^3$ Hz, it can already be sensitive to scattering cross-sections at the level of $10^{-33 (-34)}\rm{cm}^{2}$ for 1 (0.1) MeV dark matter and surpass current limits. Technological challenges include high-quality atom cluster production and injections. (2) Measure coherent neutrino-nuclei scatterings at the 0.1-1 MeV region for the first time: with $10^4$ atom guns shooting downwards CsI clusters consisting of $10^{11}$ atoms and a frequency of $10^{6}$ Hz. One can expect 10 events from MeV solar neutrinos to be observed per year. Furthermore, (3) this method can be extended to probe very low-energy neutrinos down to the eV-KeV region and may be able to detect the cosmic neutrino background, although it remains challenging.

hep-ph↗

arXiv2023

A Comparative Study of Z$^{\prime}$ mediated Charged Lepton Flavor Violation at future lepton colliders

Charged lepton flavor violation (CLFV) represents a transition between charged leptons of different generations that violates lepton flavor conservation, which is a clear signature of possible new physics beyond the standard model. By exploiting a typical example model of extra Z$^{\prime}$ gauge boson, we perform a detailed comparative study on CLFV searches at several future lepton colliders, including a 240 GeV electron-positron collider and a TeV scale muon collider. Based on detailed signal and background Monte-Carlo studies with fast detector simulations, we derive the potentials in searching for Z$^{\prime}$ mediated CLFV couplings with $e\mu$, $e\tau$ and $\mu\tau$ of different future colliders. The results are compared with the current limits set by either low-energy experiments or the high-energy LHC experiments. We find that the sensitivity of the $\tau$ related CLFV coupling strength at future lepton colliders will be significantly improved comparing with the current best constraints.

hep-ph↗

arXiv2023

Muon Beam for Neutrino CP Violation: connecting energy and neutrino frontiers

We propose here a proposal to connect neutrino and energy frontiers, by exploiting collimated muon beams for neutrino oscillations, which generate symmetric neutrino and antineutrino sources: $\mu^+\rightarrow e^+\,\bar{\nu}_{\mu}\, \nu_{e}$ and $\mu^-\rightarrow e^-\, \nu_{\mu} \,\bar{\nu}_{e}$. Interfacing with long baseline neutrino detectors such as DUNE and T2K, this experiment can be applicable to measure tau neutrino properties, and also to probe neutrino CP phase, by measuring muon electron (anti-)neutrino mixing or tau (anti-)neutrino appearance, and differences between neutrino and antineutrino rates. There are several significant benefits leading to large neutrino flux and high sensitivity on CP phase, including 1) collimated and manipulable muon beams, which lead to a larger acceptance of neutrino sources in the far detector side; 2) symmetric $\mu^+$ and $\mu^-$ beams, and thus symmetric neutrino and antineutrino sources, which make this proposal ideally useful for measuring neutrino CP violation. More importantly, $\bar{\nu}_{e,\mu}\rightarrow\bar{\nu}_\tau$ and $\nu_{e,\mu}\rightarrow \nu_\tau$, and, $\bar{\nu}_{e}\rightarrow\bar{\nu}_\mu$ and $\nu_{e}\rightarrow \nu_\mu$ oscillation signals can be collected simultaneously, with no needs for separate specific runs for neutrinos or antineutrinos. Based on a simulation of neutrino oscillation experiment, we estimate $10^4$ tau (anti-) neutrinos can be collected within 5 years which makes this proposal suitable for a brighter tau neutrino factory. Moreover, more than 7 standard deviations of sensitivity can be reached for $\dcp = |\pi/2|$, within only five ears of data taking, by combining tau and muon (anti-) neutrino appearances. With the development of a more intensive muon beam targeting future muon collider, the neutrino potential of the current proposal will surely be further improved.

hep-ph↗

arXiv2022

Application of Quantum Machine Learning in a Higgs Physics Study at the CEPC

Machine learning has blossomed in recent decades and has become essential in many fields. It significantly solved some problems in particle physics -- particle reconstruction, event classification, etc. However, it is now time to break the limitation of conventional machine learning with quantum computing. A support-vector machine algorithm with a quantum kernel estimator (QSVM-Kernel) leverages high-dimensional quantum state space to identify a signal from backgrounds. In this study, we have pioneered employing this quantum machine learning algorithm to study the $e^{+}e^{-} \rightarrow ZH$ process at the Circular Electron-Positron Collider (CEPC), a proposed Higgs factory to study electroweak symmetry breaking of particle physics. Using 6 qubits on quantum computer simulators, we optimised the QSVM-Kernel algorithm and obtained a classification performance similar to the classical support-vector machine algorithm. Furthermore, we have validated the QSVM-Kernel algorithm using 6-qubits on quantum computer hardware from both IBM and Origin Quantum: the classification performances of both are approaching noiseless quantum computer simulators. In addition, the Origin Quantum hardware results are similar to the IBM Quantum hardware within the uncertainties in our study. Our study shows that state-of-the-art quantum computing technologies could be utilised by particle physics, a branch of fundamental science that relies on big experimental data.

hep-ex↗

arXiv2022

Does Lorentz-symmetric design boost network performance in jet physics?

In the deep learning era, improving the neural network performance in jet physics is a rewarding task as it directly contributes to more accurate physics measurements at the LHC. Recent research has proposed various network designs in consideration of the full Lorentz symmetry, but its benefit is still not systematically asserted, given that there remain many successful networks without taking it into account. We conduct a detailed study on the Lorentz-symmetric design. We propose two generalized approaches for modifying a network - these methods are experimented on Particle Flow Network, ParticleNet, and LorentzNet, and exhibit a general performance gain. We also reveal that the notable improvement attributed to the "pairwise mass" feature in the network is due to its introduction of a structure that fully complies with Lorentz symmetry. We confirm that Lorentz-symmetry preservation serves as a strong inductive bias of jet physics, hence calling for attention to such general recipes in future network designs.

hep-ph↗

arXiv2022

The physics case for neutrino-neutrino collisions

Addressing the mass origin and properties of neutrinos is of strong interest to particle physics, baryogenesis and cosmology. Popular explanations involve physics beyond the standard model, for example, the dimension-5 Weinberg operator or heavy Majorana neutrinos arising from ``seesaw'' models. The current best direct limits on the electron neutrino mass, derived from nuclei beta decay or neutrinoless double beta decay processes, are at the sub-electronvolt level. Here we propose a novel neutrino neutrino collider where the neutrino beam is generated from TeV scale muon decays. Such collisions can happen between either neutrinos and anti-neutrinos, or neutrinos and neutrinos. We find that with a tiny integrated luminosity of about $10^{-5}$/fb we can already expect to observe direct neutrino anti-neutrino annihilation, $\nu\bar{\nu}\rightarrow {\rm Z}$, which also opens the door to explore neutrino related resonances $\nu\bar{\nu}\rightarrow {\rm X}$. The low luminosity requirement can accommodate a relatively large emittance muon beam. Such a device would also allow for probing heavy Majorana neutrino and effective Majorana neutrino mass through $\nu\nu\rightarrow {\rm H H}$ to a competitive level, for both electron and muon types.

hep-ph↗

arXiv2022

The physics case for a neutrino lepton collider in light of the CDF W mass measurement

We propose a neutrino lepton collider where the neutrino beam is generated from TeV scale muon decays. Such a device would allow for a precise measurement of the W mass based on single W production: nu l to W. Although it is challenging to achieve high instantaneous luminosity with such a collider, we find that a total luminosity of 0.1/fb can already yield competitive physics results. In addition to a W mass measurement, a rich variety of physics goals could be achieved with such a collider, including W boson precision measurements, heavy leptophilic gauge boson searches, and anomalous Znunu coupling searches. A neutrino lepton collider is both a novel idea in itself, and may also be a useful intermediate step, with less muon cooling required, towards the muon-muon collider already being pursued by the energy frontier community. A neutrino neutrino or neutrino proton collider may also be interesting future options for the high energy frontier.

hep-ph↗

arXiv2022

Particle Transformer for Jet Tagging

Jet tagging is a critical yet challenging classification task in particle physics. While deep learning has transformed jet tagging and significantly improved performance, the lack of a large-scale public dataset impedes further enhancement. In this work, we present JetClass, a new comprehensive dataset for jet tagging. The JetClass dataset consists of 100 M jets, about two orders of magnitude larger than existing public datasets. A total of 10 types of jets are simulated, including several types unexplored for tagging so far. Based on the large dataset, we propose a new Transformer-based architecture for jet tagging, called Particle Transformer (ParT). By incorporating pairwise particle interactions in the attention mechanism, ParT achieves higher tagging performance than a plain Transformer and surpasses the previous state-of-the-art, ParticleNet, by a large margin. The pre-trained ParT models, once fine-tuned, also substantially enhance the performance on two widely adopted jet tagging benchmarks. The dataset, code and models are publicly available at https://github.com/jet-universe/particle_transformer.

hep-ph↗

arXiv2022

An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging

Deep learning methods have been increasingly adopted to study jets in particle physics. Since symmetry-preserving behavior has been shown to be an important factor for improving the performance of deep learning in many applications, Lorentz group equivariance - a fundamental spacetime symmetry for elementary particles - has recently been incorporated into a deep learning model for jet tagging. However, the design is computationally costly due to the analytic construction of high-order tensors. In this article, we introduce LorentzNet, a new symmetry-preserving deep learning model for jet tagging. The message passing of LorentzNet relies on an efficient Minkowski dot product attention. Experiments on two representative jet tagging benchmarks show that LorentzNet achieves the best tagging performance and improves significantly over existing state-of-the-art algorithms. The preservation of Lorentz symmetry also greatly improves the efficiency and generalization power of the model, allowing LorentzNet to reach highly competitive performance when trained on only a few thousand jets. Code and models are available at \url{https://github.com/sdogsq/LorentzNet-release}.

hep-ph↗