arXiv subjects

Shen Zhang

Publications and source records attributed to Shen Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2025

LEDiT: Your Length-Extrapolatable Diffusion Transformer without Positional Encoding

Diffusion transformers (DiTs) struggle to generate images at resolutions higher than their training resolutions. The primary obstacle is that the explicit positional encodings(PE), such as RoPE, need extrapolating to unseen positions which degrades performance when the inference resolution differs from training. In this paper, We propose a Length-Extrapolatable Diffusion Transformer~(LEDiT) to overcome this limitation. LEDiT needs no explicit PEs, thereby avoiding PE extrapolation. The key innovation of LEDiT lies in the use of causal attention. We demonstrate that causal attention can implicitly encode global positional information and show that such information facilitates extrapolation. We further introduce a locality enhancement module, which captures fine-grained local information to complement the global coarse-grained position information encoded by causal attention. Experimental results on both conditional and text-to-image generation tasks demonstrate that LEDiT supports up to 4x resolution scaling (e.g., from 256x256 to 512x512), achieving better image quality compared to the state-of-the-art length extrapolation methods. We believe that LEDiT marks a departure from the standard RoPE-based methods and offers a promising insight into length extrapolation. Project page: https://shenzhang2145.github.io/ledit/

cs.CV↗

arXiv2025

Towards a 3D Transfer-based Black-box Attack via Critical Feature Guidance

Deep neural networks for 3D point clouds have been demonstrated to be vulnerable to adversarial examples. Previous 3D adversarial attack methods often exploit certain information about the target models, such as model parameters or outputs, to generate adversarial point clouds. However, in realistic scenarios, it is challenging to obtain any information about the target models under conditions of absolute security. Therefore, we focus on transfer-based attacks, where generating adversarial point clouds does not require any information about the target models. Based on our observation that the critical features used for point cloud classification are consistent across different DNN architectures, we propose CFG, a novel transfer-based black-box attack method that improves the transferability of adversarial point clouds via the proposed Critical Feature Guidance. Specifically, our method regularizes the search of adversarial point clouds by computing the importance of the extracted features, prioritizing the corruption of critical features that are likely to be adopted by diverse architectures. Further, we explicitly constrain the maximum deviation extent of the generated adversarial point clouds in the loss function to ensure their imperceptibility. Extensive experiments conducted on the ModelNet40 and ScanObjectNN benchmark datasets demonstrate that the proposed CFG outperforms the state-of-the-art attack methods by a large margin.

cs.CV↗

arXiv2025

Advantageous and disadvantageous inequality aversion can be taught through vicarious learning of others' preferences

While enforcing egalitarian social norms is critical for human society, punishing social norm violators often incurs a cost to the self. This cost looms even larger when one can benefit from an unequal distribution of resources (i.e. advantageous inequity), as in receiving a higher salary than a colleague with the identical role. In the Ultimatum Game, a classic test bed for fairness norm enforcement, individuals rarely reject (punish) such unequal proposed divisions of resources because doing so entails a sacrifice of one's own benefit. Recent work has demonstrated that observing another's punitive responses to unfairness can efficiently alter the punitive preferences of an observer. It remains an open question, however, whether such contagion is powerful enough to impart advantageous inequity aversion to individuals. Using a variant of the Ultimatum Game in which participants are tasked with responding to fairness violations on behalf of another 'Teacher' - whose aversion to advantageous (versus disadvantageous) inequity was systematically manipulated-we probe whether individuals subsequently increase their punishment unfair after experiencing fairness violations on their own behalf. In two experiments, we found individuals can acquire aversion to advantageous inequity 'vicariously' through observing (and implementing) the Teacher's preferences. Computationally, these learning effects were best characterized by a model which learns the latent structure of the Teacher's preferences, rather than a simple Reinforcement Learning account. In summary, our study is the first to demonstrate that people can swiftly and readily acquire another's preferences for advantageous inequity, suggesting in turn that behavioral contagion may be one promising mechanism through which social norm enforcement - which people rarely implement in the case of advantageous inequality - can be enhanced.

q-bio.NC↗

arXiv2025

Inverse Problems for Mean Field Games

In this book, we present a curated collection of existing results on inverse problems for Mean Field Games (MFGs), a cutting-edge and rapidly evolving field of research. Our aim is to provide fresh insights, novel perspectives, and a comprehensive foundation for future investigations into this fascinating area. MFGs, a class of differential games involving a continuum of non-atomic players, offer a powerful framework for analyzing the collective behavior of large populations of symmetric agents as the number of agents approaches infinity. This framework has proven to be an invaluable tool for quantitatively modeling the macroscopic dynamics of agents striving to minimize specific costs in complex systems, such as crowd dynamics, financial markets, traffic flows, and social networks. The study of MFGs has traditionally focused on forward problems, where the goal is to determine the equilibrium behavior of agents given a set of model parameters, such as cost functions, interaction mechanisms, and initial conditions. However, the inverse problems for MFGs -- which seek to infer these underlying parameters from observed data -- have received comparatively less attention in the literature. This book seeks to address this gap by delving into the fundamental aspects of MFG inverse problems, with a particular emphasis on issues of unique identifiability, stability, and reconstruction of unknown parameters. These problems are not only mathematically challenging but also of immense practical significance, as they enable the calibration and validation of MFG models using real-world data. This book is intended to serve as a valuable resource for researchers interested in the theory and applications of MFGs, particularly in inverse problems. Through this book, we hope to inspire further exploration and innovation in this dynamic and interdisciplinary area of study.

math.AP↗

arXiv2024

Who Saves us From Risk? Altruists Promote Cooperation in a Public Investment Game

Providing commons in the risky world is crucial for human survival, however, suffers more from the "free-riding" problem. Here, we proposed a solution that limits the access of the resource to an agent and tested its efficiency with a novel public investment game. On each trial, all group members invest in an agent responsible for a gamble and distribution. Resources are distributed evenly between non-agent group members after the agent extracts for him/herself. In 3 laboratory experiments (n = 704), we found that, as an agent, many participants extracted fewer resources than others. This altruistic distribution strategy improved cooperation levels. Furthermore, this cooperation-promoting effect could be spread to selfish agents who are in the same group as the altruistic agent and was replicated in a one-shot setting. We proposed that, when the commons are under risk, this distribution solution serves as a better alternative to the well-documented sanction solutions.

q-bio.NC↗

arXiv2024

Decoding a mean field game by the Cauchy data around its unknown stationary states

In recent years, mean field games (MFGs) have garnered considerable attention and emerged as a dynamic and actively researched field across various domains, including economics, social sciences, finance, and transportation. The inverse design and decoding of MFGs offer valuable means to extract information from observed data and gain insights into the intricate underlying dynamics and strategies of these complex physical systems. This paper presents a novel approach to the study of inverse problems in MFGs by analyzing the Cauchy data around their unknown stationary states. This study distinguishes itself from existing inverse problem investigations in three key significant aspects: Firstly, we consider MFG problems in a highly general form. Secondly, we address the technical challenge of the probability measure constraint by utilizing Cauchy data in our inverse problem study. Thirdly, we enhance existing high order linearization methods by introducing a novel approach that involves conducting linearization around non-trivial stationary states of the MFG system, which are not a-priori known. These contributions provide new insights and offer promising avenues for studying inverse problems for MFGs. By unraveling the hidden structure of MFGs, researchers and practitioners can make informed decisions, optimize system performance, and address real-world challenges more effectively.

math.AP↗

arXiv2024

Observation of hydrostatic-pressure-modulated giant caloric effect and electronic topological transition

Phase transition is a fundamental phenomenon in condensed matter physics, in which states of matter transform to each other with various critical behaviors under different conditions. The magnetic martensitic transformation features significant multi-caloric effects that benefit the solid-state cooling or heat pumping. Meanwhile, the electronic topological transition (ETT) driven by pressure has been rarely reported in martensitic systems. Here, the modulation effects of hydrostatic pressure on phase transitions in a magnetic martensitic alloy are reported. Owing to the huge volume expansion during the transition, the martensitic transition temperature is driven from 339 to 273 K by pressure within 1 GPa, resulting in highly tunable giant baro- and magneto-caloric effects (BCE and MCE) in a wide working temperature range. Interestingly, an ETT was further induced by pressure in the martensite phase, with a sudden drop of the measured saturation magnetization around 0.6 GPa. First-principles calculations reveal a sharp change in the density of states (DOS) due to the orbit shift around the Fermi level at the same pressure and reproduce the experimental observation of magnetization. Besides, the ETT is accompanied by remarkable changes in the lattice parameters and the unit-cell orthorhombicity. The study provides insight into pressure-modulated exotic phase-transition phenomena in magnetic martensitic systems.

cond-mat.mtrl-sci↗

arXiv2024

Scaling the topological transport based on an effective Weyl model

Magnetic topological semimetals are increasingly fueling interests in exotic electronic-thermal physics including thermoelectrics and spintronics. To control the transports of topological carriers in such materials becomes a central issue. However, the topological bands in real materials are normally intricate, leaving obstacles to understand the transports in a physically clear way. Parallel to the renowned effective two-band model in magnetic field scale for semiconductors, here, an effective Weyl-band model in temperature scale was developed with pure Weyl state and a few meaningful parameters for topological semimetals. Based on the model, a universal scaling was established and subsequently verified by reported experimental transports. The essential sign regularity of anomalous Hall and Nernst transports was revealed with connection to chiralities of Weyl nodes and carrier types. Upon a double-Weyl model, a concept of Berry-curvature ferrimagnetic structure, as an analogy to the real-space magnetic structure, was further proposed and well described the emerging sign reversal of Nernst thermoelectric transports in temperature scale. Our study offers a convenient tool for scaling the Weyl-fermion-related transport physics, and promotes the modulations and applications of magnetic topological materials in future topological quantum devices.

cond-mat.mes-hall↗

arXiv2024

Topological Nodal Chains and Transverse Transports in Ferromagnetic Centrosymmetric Semimetal FeIn2S4

Nodal chain semimetals protected by nonsymmorphic symmetries are distinct from Dirac and Weyl semimetals, featuring unconventional topological surface states and resulting in anomalous magnetotransport properties. Here, we reveal that the ferromagnetic FeIn2S4 is a suitable nodal chain candidate in theory. Centrosymmetric FeIn2S4 with nonsymmorphic symmetries shows half-metallicity and clean band-crossings with hourglass-type dispersion tracing out nodal lines. Owing to glide mirror symmetries, the nontrivial nodal loops form nodal chain, which is associated with the perpendicular glide mirror planes. These nodal chains are robust against spin-orbital interaction, giving rise to the coexistence of drumhead-type surface states and closed surface Fermi arcs. Moreover, the nodal loops protected by nonsymmorphic symmetry contribute to large anomalous Hall conductivity and the anomalous Nernst conductivity. Our results provide a platform to explore the intriguing topological state and transverse transport properties in magnetic system.

cond-mat.mtrl-sci↗

arXiv2024

Evolution of structure, magnetism, and electronic/thermal-transports of Ti(Cr)-substituted Fe2CrV all-d-metal Heusler ferromagnets

All-d-metal full-Heusler alloys possess superior mechanical properties and high spin polarization, which would play an important role in spintronic applications. Despite this, their electrical and thermal transport properties have not been comprehensively investigated till now. In this work, we present an analysis on the evolution of structural, magnetic and transport properties of Cr- and Ti-substituted Fe2CrV all-d-metal Heusler alloys by combining theoretical calculations and experiments. Both series of alloys crystallize in Hg2CuTi-type structure. With increasing Ti doping, the calculated total magnetic moments of Fe50Cr25V25-xTix decrease linearly. The experimental saturation magnetization is highly consistent with theoretical calculations and Slater-Pauling rule when x < 4, indicating the highly ordered atomic occupation. The magnetization and Curie temperature can be significantly tuned by altering spin polarizations and exchange interactions. The introduction of the foreign atom, Ti, results in a linear increase in residual resistivity, while electron-phonon scattering keeps relatively constant. The maximum values for electrical and thermal transport properties are observed in the stoichiometric Fe2CrV composition.

cond-mat.mtrl-sci↗

arXiv2024

Theoretical evidence of H-He demixing under Jupiter and Saturn conditions

The immiscibility of hydrogen-helium mixture under the temperature and pressure conditions of planetary interiors is crucial for understanding the structures of gas giant planets (e.g., Jupiter and Saturn). While the experimental probe at such extreme conditions is challenging, theoretical simulation is heavily relied in an effort to unravel the mixing behavior of hydrogen and helium. Here we develop a method via a machine learning accelerated molecular dynamics simulation to quantify the physical separation of hydrogen and helium under the conditions of planetary interiors. The immiscibility line achieved with the developed method yields substantially higher demixing temperatures at pressure above 1.5 Mbar than earlier theoretical data, but matches better to the experimental estimate. Our results suggest a possibility that H-He demixing takes place in a large fraction of the interior radii of Jupiter and Saturn, i.e., 27.5% in Jupiter and 48.3% in Saturn. This indication of an H-He immiscible layer hints at the formation of helium rain and offers a potential explanation for the decrease of helium in the atmospheres of Jupiter and Saturn.

physics.comp-ph↗

arXiv2024

Stability of Persistent Path Diagrams

In real-world systems, the relationships and connections between components are highly complex. Real systems are often described as networks, where nodes represent objects in the system and edges represent relationships or connections between nodes. With the deepening of research, networks have been endowed with richer structures, such as directed edges, edge weights, and even hyperedges involving multiple nodes. Persistent homology is an algebraic method for analyzing data. It helps us understand the intrinsic structure and patterns of data by tracking the death and birth of topological features at different scale parameters.The original persistent homology is not suitable for directed networks. However, the introduction of path homology established on digraphs solves this problem. This paper studies complex networks represented as weighted digraphs or edge-weighted path complexes and their persistent path homology. We use the homotopy theory of digraphs and path complexes, along with the interleaving property of persistent modules and bottleneck distance, to prove the stability of persistent path diagram with respect to weighted digraphs or edge-weighted path complexes. Therefore, persistent path homology has practical application value.

math.AT↗

arXiv2024

Radiative Thermal Transistor

Developing thermal analogues of field-effect transistor could open the door to a low-power and even zero-power communication technology working with heat rather than electricity. These solid-sate devices could also find many applications in the field of active thermal management in numerous technologies (microelectronic, building science, energy harvesting,conversion,...). Recent theoretical works has suggested that a photonic transistor made with three terminals can in principle be used to switch, modulate, and even amplify heat flux through exchange of thermal photons. Here, we report an experimental demonstration of thermal transistor effect using a non-contact system composed by a temperature-controlled metal-insulator-based material interacting in far-field regime with two blackbodies held at two different temperatures. We demonstrate that, with a tiny change in the temperature of the active layer, the heat flux received by the cold blackbody can be drastically modified. An amplification parameter of heat flux over 20 is reported.

physics.app-ph↗

arXiv2024

HiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Models

Diffusion models have become a mainstream approach for high-resolution image synthesis. However, directly generating higher-resolution images from pretrained diffusion models will encounter unreasonable object duplication and exponentially increase the generation time. In this paper, we discover that object duplication arises from feature duplication in the deep blocks of the U-Net. Concurrently, We pinpoint the extended generation times to self-attention redundancy in U-Net's top blocks. To address these issues, we propose a tuning-free higher-resolution framework named HiDiffusion. Specifically, HiDiffusion contains Resolution-Aware U-Net (RAU-Net) that dynamically adjusts the feature map size to resolve object duplication and engages Modified Shifted Window Multi-head Self-Attention (MSW-MSA) that utilizes optimized window attention to reduce computations. we can integrate HiDiffusion into various pretrained diffusion models to scale image generation resolutions even to 4096x4096 at 1.5-6x the inference speed of previous methods. Extensive experiments demonstrate that our approach can address object duplication and heavy computation issues, achieving state-of-the-art performance on higher-resolution image synthesis tasks.

cs.CV↗

arXiv2024

Samsung Research China-Beijing at SemEval-2024 Task 3: A multi-stage framework for Emotion-Cause Pair Extraction in Conversations

In human-computer interaction, it is crucial for agents to respond to human by understanding their emotions. Unraveling the causes of emotions is more challenging. A new task named Multimodal Emotion-Cause Pair Extraction in Conversations is responsible for recognizing emotion and identifying causal expressions. In this study, we propose a multi-stage framework to generate emotion and extract the emotion causal pairs given the target emotion. In the first stage, Llama-2-based InstructERC is utilized to extract the emotion category of each utterance in a conversation. After emotion recognition, a two-stream attention model is employed to extract the emotion causal pairs given the target emotion for subtask 2 while MuTEC is employed to extract causal span for subtask 1. Our approach achieved first place for both of the two subtasks in the competition.

cs.CL↗

arXiv2024

Simultaneously recovering running cost and Hamiltonian in Mean Field Games system

We propose and study several inverse problems for the mean field games (MFG) system in a bounded domain. Our focus is on simultaneously recovering the running cost and the Hamiltonian within the MFG system by the associated boundary observation. There are several technical novelties that make the study intriguing and challenging. First, the MFG system couples two nonlinear parabolic PDEs with one moving forward and the other one moving backward in time. Second, there is a probability density constraint on the population distribution of the agents. Third, the simultaneous recovery of two coupling factors within the MFG system is technically far from being trivial. Fourth, we consider both cases that the running cost depends on the population density locally and non-locally, and the two cases present different technical challenges for the inverse problem study. We develop two mathematical strategies that can ensure the probability constraint as well as effectively tackle the inverse problems, which are respectively termed as high-order variation and successive linearisation. In particular, the high-order variation method is new to the literature, which demonstrates a novel concept to examine the inverse problems by non-negative inputs only. We believe the methods developed can find applications to inverse problems in other contexts.

math.OC↗

arXiv2024

Inverse boundary problem for a mean field game system with probability density constraint

By following the study in [24], we consider an inverse boundary problem for the mean field game system where a probability density constraint is enforced on the game agents. That is, we consider the case that reflective boundary conditions are enforced and hence the population distribution of the game agents should be treated as a probability measure which preserves both positivity and the total population. This poses significant challenges for the corresponding inverse problems in constructing suitable ``probing modes" which should fulfill such a probability density constraint. We develop an effective scheme in tackling such a case which is new to the literature.

math.AP↗

arXiv2023

Full-scale ab initio simulations of laser-driven atomistic dynamics

The coupling of excited states and ionic dynamics is the basic and challenging point for the materials response at extreme conditions. In laboratory, the intense laser produces transient nature and complexity with highly nonequilibrium states, making it extremely difficult and interesting for both experimental measurements and theoretical methods. With the inclusion of laser-excited states, we extended ab initio method into the direct simulations of whole laser-driven microscopic dynamics from solid to liquid. We constructed the framework of combining the electron-temperaturedependent deep neural network potential energy surface with hybrid atomistic-continuum approach, controlling non-adiabatic energy exchange and atomistic dynamics, which enables consistent interpretation of experimental data. By large scale ab inito simulations, we demonstrate that the nonthermal effects introduced by hot electrons play a dominant role in modulating the lattice dynamics, thermodynamic pathway, and structural transformation. We highlight that the present work provides a path to realistic computational studies of laser-driven processes, thus bridging the gap between experiments and simulations.

physics.comp-ph↗