arXiv subjects

Jin Sun

Publications and source records attributed to Jin Sun.

At least 55 records · Page 3Linked to original sources

arXiv2024

Maximal flavor violating $U(1)_{L_μ-L_τ}$ model with singly-charged scalar

The $U(1)_{L_μ-L_τ}$ $Z'$ model has emerged as a promising candidate to address the longstanding muon $(g-2)_μ$ anomaly. Flavor-conserving $Z'$ interactions are subject to stringent constraints from the neutrino trident process and NA64$μ$ experiments, which limit the $Z'$ mass to $m_{Z'}<40$ MeV. To circumvent these constraints, flavor-conserving $Z'$ interactions can be converted into maximal flavor-violating interactions through a discrete exchange symmetry. Maximal flavor-violating $Z'$ interactions contribute to $τ\toμν\barν$ via the neutral current, yet the parameter space for this process conflicts with the $(g-2)_μ$ allowed regions. To resolve this conflict, we propose introducing a singly-charged scalar that mediates via charged current interactions. This scalar is anticipated to produce a negative contribution to the lepton $(g-2)_l$ while concurrently inducing the tau decay $τ\toμν\barν$. Three distinct scenarios arise from the introduction of singly-charged scalars: the $\mathcal{O}_{LL}$ operator, driven by weak singlet and triplet, and the $\mathcal{O}_{LR}$ operator, driven by weak doublet. Our analysis of the phenomenology in these three cases reveals that the tension between the muon $(g-2)_μ$ anomaly and $τ\toμν\barν$ for large $Z'$ masses can be effectively alleviated only by the singly-charged scalars from the weak triplet, whereas the singlet and doublet scenarios fall short. Furthermore, the singly-charged scalars from the weak triplet offer an additional explanation for the electron $(g-2)_e$ anomaly. Our findings indicate that weak triplets could play a crucial role in $Z'$ models, potentially providing valuable insights for future research into $U(1)$ frameworks.

hep-ph↗

arXiv2024

Neural Gaffer: Relighting Any Object via Diffusion

Single-image relighting is a challenging task that involves reasoning about the complex interplay between geometry, materials, and lighting. Many prior methods either support only specific categories of images, such as portraits, or require special capture conditions, like using a flashlight. Alternatively, some methods explicitly decompose a scene into intrinsic components, such as normals and BRDFs, which can be inaccurate or under-expressive. In this work, we propose a novel end-to-end 2D relighting diffusion model, called Neural Gaffer, that takes a single image of any object and can synthesize an accurate, high-quality relit image under any novel environmental lighting condition, simply by conditioning an image generator on a target environment map, without an explicit scene decomposition. Our method builds on a pre-trained diffusion model, and fine-tunes it on a synthetic relighting dataset, revealing and harnessing the inherent understanding of lighting present in the diffusion model. We evaluate our model on both synthetic and in-the-wild Internet imagery and demonstrate its advantages in terms of generalization and accuracy. Moreover, by combining with other generative methods, our model enables many downstream 2D tasks, such as text-based relighting and object insertion. Our model can also operate as a strong relighting prior for 3D tasks, such as relighting a radiance field.

cs.CV↗

arXiv2024

SU(3) symmetry analysis in charmed baryon two body decays with penguin diagram contribution

An increasing number of experimental measurements from the BESIII, Belle, and Belle-II collaborations encourage investigations into charmed baryon two-body decay processes. By including contributions from the penguin diagrams that are ignored in previous studies, we perform a global analysis with SU(3) flavor symmetry. Assuming all form factors are real, we achieve a remarkable minimal $χ^{2}/d.o.f = 0.788$ and find that the contribution of the amplitude proportional to $V_{cb}^*V_{ub}$ is of the order $\sim 0.01$, comparable with the contribution of the tree-level diagram. Additionally, by using the KPW theorem to reduce the number of amplitudes from 13 to 7 in the leading contribution, it becomes possible to consider the complex form factor case for the leading IRA amplitude in the global analysis. However, the analysis of complex form factors significantly conflicts with the experimental data $Br(Ξ_c^0\toΞ^-π^+)$, and by excluding this data, $χ^2/d.o.f$ is reduced from 5.95 to 1.19. Although the analysis of complex form factors shows a significant central value of the penguin diagram contribution, the large errors from the corresponding form factors make it a challenge to precisely determine its true contribution. Consequently, the direct CP violation in decay processes is predicted to be approximately zero. With more data in future experiments, the penguin diagram contribution with the amplitude proportional to $V_{cb}^*V_{ub}$ will be precisely determined, allowing for a more accurate prediction of CP violation. Our analysis necessitates further theoretical investigations and experimental measurements in the future.

hep-ph↗

arXiv2024

Caterpillar: A Pure-MLP Architecture with Shifted-Pillars-Concatenation

Modeling in Computer Vision has evolved to MLPs. Vision MLPs naturally lack local modeling capability, to which the simplest treatment is combined with convolutional layers. Convolution, famous for its sliding window scheme, also suffers from this scheme of redundancy and lower parallel computation. In this paper, we seek to dispense with the windowing scheme and introduce a more elaborate and parallelizable method to exploit locality. To this end, we propose a new MLP module, namely Shifted-Pillars-Concatenation (SPC), that consists of two steps of processes: (1) Pillars-Shift, which generates four neighboring maps by shifting the input image along four directions, and (2) Pillars-Concatenation, which applies linear transformations and concatenation on the maps to aggregate local features. SPC module offers superior local modeling power and performance gains, making it a promising alternative to the convolutional layer. Then, we build a pure-MLP architecture called Caterpillar by replacing the convolutional layer with the SPC module in a hybrid model of sMLPNet. Extensive experiments show Caterpillar's excellent performance on both small-scale and ImageNet-1k classification benchmarks, with remarkable scalability and transfer capability possessed as well. The code is available at https://github.com/sunjin19126/Caterpillar.

cs.CV↗

arXiv2024

Curvature Estimate of Nodal Sets of Harmonic Functions in the Plane

In this paper, we study curvature estimates for nodal sets of harmonic functions in the plane. We prove that at any point $p$, the curvature of each nodal curve of any harmonic function $u$ is upper bounded by $$\left|{κ(u)(p)}\right|\leq \frac{4(n+1)}{nr}\cos nα_0,$$ where $u$ has only $n$ nodal curves in $B_r(p)$ intersecting at $p$, and $α_0=0$ for odd $n$ or $α_0=\fracπ{2n(n+1)}$ for even $n$. This result is sharp for all $n\geq 1$. In extreme cases, $u$ can be given by the Poisson extension of Dirac measure and its derivatives. Moreover, the curvature of any nodal curve is uniformly upper bounded at every point in the nodal set of $u$ in a small neighborhood $B_{cr}(p)$, where $c<1$ depends only on $n$. Furthermore, with the frequency tool, we prove that the area of the positive part and the negative part of $u$ have a uniform lower bound, which depends only on the number of nodal domains in $B_r(p)$.

math.AP↗

arXiv2024

A Benchmark Dataset for Multimodal Prediction of Enzymatic Function Coupling DNA Sequences and Natural Language

Predicting gene function from its DNA sequence is a fundamental challenge in biology. Many deep learning models have been proposed to embed DNA sequences and predict their enzymatic function, leveraging information in public databases linking DNA sequences to an enzymatic function label. However, much of the scientific community's knowledge of biological function is not represented in these categorical labels, and is instead captured in unstructured text descriptions of mechanisms, reactions, and enzyme behavior. These descriptions are often captured alongside DNA sequences in biological databases, albeit in an unstructured manner. Deep learning of models predicting enzymatic function are likely to benefit from incorporating this multi-modal data encoding scientific knowledge of biological function. There is, however, no dataset designed for machine learning algorithms to leverage this multi-modal information. Here we propose a novel dataset and benchmark suite that enables the exploration and development of large multi-modal neural network models on gene DNA sequences and natural language descriptions of gene function. We present baseline performance on benchmarks for both unsupervised and supervised tasks that demonstrate the difficulty of this modeling objective, while demonstrating the potential benefit of incorporating multi-modal data types in function prediction compared to DNA sequences alone. Our dataset is at: https://hoarfrost-lab.github.io/BioTalk/.

q-bio.GN↗

arXiv2024

Ramsey-type results on parameters related to domination

The following inequality chain $$ ir(G)\le γ(G)\le i(G)\le α(G) \le Γ(G) \le I\!R(G)$$ is known as a domination chain, where $ir(G), γ(G), i(G), α(G), Γ(G)$, and $I\!R(G)$ are the lower irredundance number, the domination number, the independence domination number, the independence number, the upper domination number and the upper irredundance number of $G$, respectively. The Ramsey-type problem seeks to characterize the family $\mathcal H$ of graphs such that every $\mathcal H$-free graph $G$ has a bounded parameter $μ$. The classical Ramsey's theorem states that every $\{K_n, E_n\}$-free graph has a bounded number of vertices. Furuya (Discrete Math.Theor 2018) characterized $\mathcal H$ such that every connected $\mathcal H$-free graph $G$ has a bounded domination number. The characterization of the graph family $\mathcal H$ for which every connected $\mathcal H$-free graph $G$ has a bounded independence number was due to Choi, Furuya, Kim, Park~(Discrete math. 2020) and Chiba, Furuya (Electron. J. Combin., 2022). In this paper, we further characterize $\mathcal H$ such that every connected $\mathcal H$-free graph $G$ has bounded $μ(G)$ for $μ$ belonging to the set $\{ir(G), i(G), Γ(G), \text{IR}(G)\}$. This completes the characterization of $\mathcal H$ for which every connected $\mathcal H$-free graph $G$ has bounded $μ(G)$ for $μ(G)$ along the domination chain. Additionally, we characterize $\mathcal H$ such that every connected $\mathcal H$-free graph $G$ has bounded $μ(G)$ for $μ$ related to the domination number. Specifically, we consider the parameters $O\!I\!R(G)$, $I\!S(G)$, or $I\!R\!S(G)\}$, where $O\!I\!R(G)$, $I\!S(G)$, and $I\!R\!S(G)$ are the open irredundance number, the independence saturation number, and the irredundance saturation number of graph $G$, respectively.

math.CO↗

arXiv2024

A Ramsey-type theorem on deficiency

Ramsey's Theorem states that a graph $G$ has bounded order if and only if $G$ contains no complete graph $K_n$ or empty graph $E_n$ as its induced subgraph. The Gyárfás-Sumner conjecture says that a graph $G$ has bounded chromatic number if and only if it contains no induced subgraph isomorphic to $K_n$ or a tree $T$. The deficiency of a graph is the number of vertices that cannot be covered by a maximum matching. In this paper, we prove a Ramsey type theorem for deficiency, i.e., we characterize all the forbidden induced subgraphs for graphs $G$ with bounded deficiency. As an application, we answer a question proposed by Fujita, Kawarabayashi, Lucchesi, Ota, Plummer and Saito (JCTB, 2006).

math.CO↗

arXiv2024

MVSA-Net: Multi-View State-Action Recognition for Robust and Deployable Trajectory Generation

The learn-from-observation (LfO) paradigm is a human-inspired mode for a robot to learn to perform a task simply by watching it being performed. LfO can facilitate robot integration on factory floors by minimizing disruption and reducing tedious programming. A key component of the LfO pipeline is a transformation of the depth camera frames to the corresponding task state and action pairs, which are then relayed to learning techniques such as imitation or inverse reinforcement learning for understanding the task parameters. While several existing computer vision models analyze videos for activity recognition, SA-Net specifically targets robotic LfO from RGB-D data. However, SA-Net and many other models analyze frame data captured from a single viewpoint. Their analysis is therefore highly sensitive to occlusions of the observed task, which are frequent in deployments. An obvious way of reducing occlusions is to simultaneously observe the task from multiple viewpoints and synchronously fuse the multiple streams in the model. Toward this, we present multi-view SA-Net, which generalizes the SA-Net model to allow the perception of multiple viewpoints of the task activity, integrate them, and better recognize the state and action in each frame. Performance evaluations on two distinct domains establish that MVSA-Net recognizes the state-action pairs under occlusion more accurately compared to single-view MVSA-Net and other baselines. Our ablation studies further evaluate its performance under different ambient conditions and establish the contribution of the architecture components. As such, MVSA-Net offers a significantly more robust and deployable state-action trajectory generation compared to previous methods.

cs.CV↗

arXiv2024

ALDM-Grasping: Diffusion-aided Zero-Shot Sim-to-Real Transfer for Robot Grasping

To tackle the "reality gap" encountered in Sim-to-Real transfer, this study proposes a diffusion-based framework that minimizes inconsistencies in grasping actions between the simulation settings and realistic environments. The process begins by training an adversarial supervision layout-to-image diffusion model(ALDM). Then, leverage the ALDM approach to enhance the simulation environment, rendering it with photorealistic fidelity, thereby optimizing robotic grasp task training. Experimental results indicate this framework outperforms existing models in both success rates and adaptability to new environments through improvements in the accuracy and reliability of visual grasping actions under a variety of conditions. Specifically, it achieves a 75\% success rate in grasping tasks under plain backgrounds and maintains a 65\% success rate in more complex scenarios. This performance demonstrates this framework excels at generating controlled image content based on text descriptions, identifying object grasp points, and demonstrating zero-shot learning in complex, unseen scenarios.

cs.RO↗

arXiv2024

Light baryon in three quark picture light front approach and its application: hyperon weak radiative decays

Motivated by recent experimental data on $Σ^+\to pγ$ at BESIII, we investigate a class of hyperon weak radiative decays. To estimate these processes, in our research, we employ a new type of light-front quark model with a three-quark picture for octet baryons. In the three-quark picture, with the use of $SU(3)_f$ and spin symmetries, we present a general form of the light front wave function for each octet baryon. By including contributions from the penguin diagram and W exchange diagram, we perform a complete calculation on the branching ratios ($Br$) and the asymmetry parameter ($α$) for hyperon weak radiative decay processes. Our results are helpful for discovering additional hyperon weak radiative decay processes in experimental facilities, and our research will promote the theoretical study of baryons.

hep-ph↗

arXiv2023

WiSegRT: Dataset for Site-Specific Indoor Radio Propagation Modeling with 3D Segmentation and Differentiable Ray-Tracing

The accurate modeling of indoor radio propagation is crucial for localization, monitoring, and device coordination, yet remains a formidable challenge, due to the complex nature of indoor environments where radio can propagate along hundreds of paths. These paths are resulted from the room layout, furniture, appliances and even small objects like a glass cup. They are also influenced by the object material and surface roughness. Advanced machine learning (ML) techniques have the potential to take such non-linear and hard-to-model factors into consideration. However, extensive and fine-grained datasets are urgently required. This paper presents WiSegRT, an open-source dataset for indoor radio propagation modeling. Generated by a differentiable ray tracer within the segmented 3-dimensional (3D) indoor environments, WiSegRT provides site-specific channel impulse responses for each grid point relative to the given transmitter location. We expect WiSegRT to support a wide-range of applications, such as ML-based channel prediction, accurate indoor localization, radio-based object detection, wireless digital twin, and more.

cs.IT↗

arXiv2023

Dark photon effects with the kinetic and mass mixing in Z boson decay processes

Motivated by the most recent measurement of tau polarization in $Z\to τ^+τ^-$ by CMS, we have introduced a new $U(1)_X$ gauge boson field X, which can have renormalizable kinetic mixing with the standard model $U(1)_Y$ gauge boson field Y. In addition to the kinetic mixing of the dark photon, denoted as $σ$, there may also be mass mixing introduced by the additional Higgs doublet with a vacuum expectation value (vev) participating in $U(1)_X$ and electroweak symmetry breaking simultaneously. The interaction of the Z boson with the SM leptons is modified by the introduction of the mixing ratio parameter $ε$, which quantifies the magnitude of both the mass and kinetic mixing of the dark photon. Initially, we use the tau lepton as an example to explore the Z boson phenomenology of the dark photon model with both kinetic and mass mixing. The goal is to determine the allowed parameter regions by taking into account constraints from the vector and axial-vector couplings $g_{V,A}^τ$, the decay branching ratio $Br(Z\to τ^- τ^+)$ and tau lepton polarization in $Z\to τ^-τ^+$. We found that the mixing ratio plays important role in the Z boson features by choosing different $ε$ values. Furthermore, we aim to generalize our analysis from the tau-lepton case to include all fermions by conducting global fits. This allows us to identify viable regions by incorporating relevant fermion constraints and the W/Z mass ratio. Correspondingly, we obtain the fit results with the kinetic mixing parameter $σ=0.074\pm0.021$, mixing ratio $ε=-1.37\pm0.46$, and dark photon mass $m_X=275\pm39$ GeV. Our global analysis indicates a preference for a dark photon mass larger than $m_Z$.

hep-ph↗

arXiv2023

Black-box Backdoor Defense via Zero-shot Image Purification

Backdoor attacks inject poisoned samples into the training data, resulting in the misclassification of the poisoned input during a model's deployment. Defending against such attacks is challenging, especially for real-world black-box models where only query access is permitted. In this paper, we propose a novel defense framework against backdoor attacks through Zero-shot Image Purification (ZIP). Our framework can be applied to poisoned models without requiring internal information about the model or any prior knowledge of the clean/poisoned samples. Our defense framework involves two steps. First, we apply a linear transformation (e.g., blurring) on the poisoned image to destroy the backdoor pattern. Then, we use a pre-trained diffusion model to recover the missing semantic information removed by the transformation. In particular, we design a new reverse process by using the transformed image to guide the generation of high-fidelity purified images, which works in zero-shot settings. We evaluate our ZIP framework on multiple datasets with different types of attacks. Experimental results demonstrate the superiority of our ZIP framework compared to state-of-the-art backdoor defense baselines. We believe that our results will provide valuable insights for future defense methods for black-box models. Our code is available at https://github.com/sycny/ZIP.

cs.CV↗

arXiv2023

Map2Schedule: An End-to-End Link Scheduling Method for Urban V2V Communications

Urban vehicle-to-vehicle (V2V) link scheduling with shared spectrum is a challenging problem. Its main goal is to find the scheduling policy that can maximize system performance (usually the sum capacity of each link or their energy efficiency). Given that each link can experience interference from all other active links, the scheduling becomes a combinatorial integer programming problem and generally does not scale well with the number of V2V pairs. Moreover, link scheduling requires accurate channel state information (CSI), which is very difficult to estimate with good accuracy under high vehicle mobility. In this paper, we propose an end-to-end urban V2V link scheduling method called Map2Schedule, which can directly generate V2V scheduling policy from the city map and vehicle locations. Map2Schedule delivers comparable performance to the physical-model-based methods in urban settings while maintaining low computation complexity. This enhanced performance is achieved by machine learning (ML) technologies. Specifically, we first deploy the convolutional neural network (CNN) model to estimate the CSI from street layout and vehicle locations and then apply the graph embedding model for optimal scheduling policy. The results show that the proposed method can achieve high accuracy with much lower overhead and latency.

cs.NI↗

arXiv2023

Towards Artificial General Intelligence (AGI) in the Internet of Things (IoT): Opportunities and Challenges

Artificial General Intelligence (AGI), possessing the capacity to comprehend, learn, and execute tasks with human cognitive abilities, engenders significant anticipation and intrigue across scientific, commercial, and societal arenas. This fascination extends particularly to the Internet of Things (IoT), a landscape characterized by the interconnection of countless devices, sensors, and systems, collectively gathering and sharing data to enable intelligent decision-making and automation. This research embarks on an exploration of the opportunities and challenges towards achieving AGI in the context of the IoT. Specifically, it starts by outlining the fundamental principles of IoT and the critical role of Artificial Intelligence (AI) in IoT systems. Subsequently, it delves into AGI fundamentals, culminating in the formulation of a conceptual framework for AGI's seamless integration within IoT. The application spectrum for AGI-infused IoT is broad, encompassing domains ranging from smart grids, residential environments, manufacturing, and transportation to environmental monitoring, agriculture, healthcare, and education. However, adapting AGI to resource-constrained IoT settings necessitates dedicated research efforts. Furthermore, the paper addresses constraints imposed by limited computing resources, intricacies associated with large-scale IoT communication, as well as the critical concerns pertaining to security and privacy.

cs.AI↗

arXiv2023

Some Ramsey-type results

The Ramsey's theorem says that a graph with sufficiently many vertices contains a clique or stable set with many vertices. Now we attach some parameter to every vertex, such as degree. Consider the case a graph with sufficiently many vertices of large degree, we can get the realted Ramsey-type result. The Ramsey's theorem of connected version says that every connected graph with sufficiently many vertices contains an induced path, clique or star with many vertices. Now we require the vertex is non-trivial, i.e. the parameter of this vertex is non-trivial, such as $\operatorname{deg}(v)\ge 2$. A connected graph with sufficiently many non-trivial vertices must contain some special induced subgraph. We also get the non-connected version of this Ramsey-type result as a corollary.

math.CO↗

arXiv2023

Muon collider signatures for a $Z'$ with a maximal $μ-τ$ coupling in $U(1)_{L_μ-L_τ}$

The gauged $U(1)_{L_μ- L_τ}$ model is a candidate model for explaining the muon g-2 anomaly because the $Z'$ in the model has a natural normal coupling to muon. Due to other experimental data constraints the viable mass range for the usual $Z'$ is constrained to be lower than a few hundred MeV. It has been shown that if the $Z'$ has a maximal off-diagonal mixing, $(\bar μγ^μτ+ \bar τγ^μμ) Z'_μ$, a large mass for $Z'$ is possible. This class of models has a very interesting signature for detection, such as $μ^-μ^+ \to τ^- τ^+$ pair, $μ^- μ^+ \to μ^\pmμ^\pm τ^\mp τ^\mp$ at a muon collider. In this work we study in detail these processes. We find that the in the parameter space solving the muon g-2 anomaly, t-channel $τ^- τ^+$ pair production can easily be distinguished at more than 5$σ$ level from the s-channel production as that predicted in the standard model. The smoking gun signature of doubly same sign $μ^\pm μ^\pm + τ^\mpτ^\mp$ pairs production can have a 5$σ$ sensitivity, at a muon collider of 3 TeV with $\mathcal{O}$($fb^{-1}$) luminosity.

hep-ph↗