arXiv subjects

Ying Zhao

Publications and source records attributed to Ying Zhao.

At least 19 recordsLinked to original sources

arXiv2026

Message Order Selects Opposite Collective Opinions on Laplacian-Cospectral Networks

Receiving the same messages in different orders can change an agent's judgement. We construct a network model in which this local sequence effect determines the final opinion of an entire network. Each agent has a two-level quantum-like state with an expressed opinion and an auxiliary context coordinate that can affect the next update. Two valid update rules of equal strength leave the neutral state unchanged and add no positive or negative opinion bias. Nevertheless, reversing their order gives the exact opinions +0.01 and -0.01 from the same initial state. We place four ordered pairs so that their positive and negative contributions sum to zero but occupy different labelled nodes. After this single input, all cases follow the same nonlinear network rule, which preserves valid states and admits two stable consensuses. We compare two 16-node networks whose complete Laplacian spectra are identical. With every local ingredient fixed, they converge to opposite consensuses. A separate local calculation identifies a specific overlap between the labelled pulse and Laplacian eigenspace projectors that sets the leading displacement of the basin boundary. Rigorous interval arithmetic proves the finite registered example independently. The construction connects message order to collective selection and shows why Laplacian eigenvalues alone do not determine which opinion wins.

physics.soc-ph↗

arXiv2026

Booster-based beam recycling for swap-out injection at the High Energy Photon Source

Fourth-generation synchrotron light sources employ ultralow-emittance storage rings with stringent injection requirements. On-axis swap-out injection alleviates the dependence on storage-ring dynamic aperture, but high-charge operation requires an efficient injector architecture capable of producing high-charge replacement bunches. This paper presents the accelerator physics design and performance analysis of a booster-based beam-recycling swap-out injection scheme implemented at the High Energy Photon Source (HEPS). In this approach, the full-energy booster serves as both an injector and a high-energy accumulator. An extracted storage-ring bunch is returned to the booster, merged with a low-charge bunch previously injected from the linac and accelerated to full energy. Following high-energy damping, the merged bunch is reinjected into the original storage-ring bucket. The scheme avoids the need for a dedicated accumulator ring while enabling high-charge bunch replacement. The recycling scheme was commissioned through staged machine studies. Full recycling-chain simulations, commissioning studies, and measured performance analysis are presented. The measured results characterize the recycling operation and quantify the transmission efficiency and performance limitations of the complete recycling loop. These results demonstrate the feasibility of the booster-based beam-recycling architecture and establish its operational basis for high-charge swap-out injection in future fourth-generation synchrotron light sources.

physics.acc-ph↗

arXiv2026

Same Resident Strains, Different Attractors: Opposite Local Growth Signs for a Rare Third Strain

Most models assess whether a newly introduced pathogen strain can grow when rare by testing it against a steady endemic background. Resident strains, however, can have more than one long-term behavior under the same parameters. They may settle to a steady state or continue through recurring outbreaks, leaving different fractions of hosts with different infection histories. We study this possibility in a stylized immune-history model. For two resident strains, we identify a parameter point where a stable endemic equilibrium and a stable outbreak cycle are both locally attracting. Numerical continuation maps a Hopf branch and a branch of folds of periodic orbits that bound a candidate bistability region, organized locally by a generalized Hopf (Bautin) point. We then embed the resident model exactly in a 20-state, three-strain system. When strain 3 infectiousness and total removal are common across four host-history classes, rare growth is determined by a weighted sum of the corresponding uninfected host fractions. The equilibrium and cycle yield different weighted sums, so one fixed strain 3 parameter set grows near one resident attractor but declines near the other. A second parameter set gives the reverse ordering, and both patterns persist over open regions of invader parameter space. Sixty full-model simulations with progressively smaller inocula reproduce these linear predictions across cycle phases and numerical solvers. Thus, knowing which resident strains are present may not suffice to predict initial invasion; the realized resident attractor can also matter. These results concern local growth from rarity, not eventual establishment, long-term coexistence, or post-invasion dynamics.

math.DS↗

arXiv2026

Giant exciton effects and magneto-excitonic coupling in V4S9X4 2D magnetic semiconductors

Room-temperature spin-optoelectronic devices require a combination of robust ferromagnetism and giant exciton binding, a pairing mutually exclusive in conventional semiconductors due to magnetic localization that screens excitons. Cluster-assembled V4S9X4 (X = F, Cl, Br and I) monolayers overcome this bottleneck via a hierarchical design, that is, intra-cluster localized states host both local magnetic moments and strong electron-hole interactions, while inter-cluster coupling mediates long-range ferromagnetism. Remarkably, these two-dimensional semiconductors exhibit intrinsic ferromagnetism with Curie temperature up to 507.6 K. As a prototype, V4S9Br4 monolayer possesses a giant exciton binding energy of 1.85 eV. Its lowest exciton is a dark state (DI) with a radiative lifetime of 1.20 ns, whereas the first bright exciton (BI) exhibits an ultrafast radiative decay of 86.87 ps. This stark lifetime contrast enables simultaneous ultrafast optical response and long-lived spin information storage. Most notably, switching between ferromagnetic and antiferromagnetic order allows for wide-range tuning of exciton lifetime, with the giant binding energy remaining nearly intact. Our findings establish cluster assembly as a powerful paradigm for designing next-generation spin-photonic and quantum information devices operating at room temperature.

cond-mat.mtrl-sci↗

arXiv2026

Reliability-Contagion Feasibility in LLM Multi-Agent Networks

Communication allows large language model agents to pool evidence, but it also creates paths along which an erroneous claim can spread. We formulate a correction-aware network model that tracks susceptible, exposed, infectious, and corrected agents and derive its early-invasion condition for heterogeneous communication networks. We then couple this propagation model to an analytic majority-vote benchmark in which a clean-task reliability target imposes a minimum connectivity requirement. Under fixed exposure per communication edge, reliability and error control impose opposing graph constraints. We characterize when their intersection is empty and when it contains an intermediate connectivity range, and identify regular graphs that attain the smallest invasion factor within the reliable graph class when such graphs exist. Under a fixed sender budget, the homogeneous first-order threshold is independent of network density, showing that the communication-budget convention determines whether added edges increase early propagation risk. Finite-network simulations on 21,000 trajectories illustrate these directional predictions. A controlled grok-4.3 experiment then evaluates three six-node topologies on 36 new closed-world tasks, with a balanced 12-task subset continued to full cascades. Mean first-generation offspring increased from 0.667 to 1.333 and 1.667 as degree increased from 2 to 4 and 5, while the adoption fraction among exposed neighbours remained 0.333. Mean non-seed erroneous adoption in the full-cascade subset was 0.200, 0.333, and 0.333. Together, these results provide a tractable basis for selecting connectivity under explicit reliability and propagation constraints.

cs.MA↗

arXiv2026

Anti-scrambling and euclidean folds from observer correlators in de Sitter space

The de Sitter horizon behaves in a qualitatively different way from a black hole horizon, which poses a challenge to any attempt to develop a fundamental quantum description of de Sitter space. In this paper we gather some ``data'' on this problem using gravitational calculations, seeing that they lead to an ``anti-scrambling'' phenomenon that is contrary to the behavior of standard many-body quantum systems. We organize our discussion in terms of correlation functions computed on the worldline of an observer living in the spacetime, with the two-point function looking like a conventional thermal correlator but the four-point function showing anti-scrambling. We propose that anti-scrambling can be realized in a quantum system whose Hamiltonian is bounded from both above and below using correlators that are folded in Euclidean time.

hep-th↗

arXiv2026

DiffLens: A Visualization System to Explore Local Differences in Graph Sampling

Graph sampling techniques have been widely used to simplify network computation and visualization, which also results in inevitable differences between the sampled networks and the original networks in terms of nodes, edges and structures. Investigating such differences can inform graph sampling technique users of the pros and cons of different techniques and select the appropriate one, and can also help graph sampling developers evaluate their own technique. However, there are still no systematic ways to achieve such a goal. This paper fills this research gap by first proposing systematic and generic quantitative measures to quantify three categories of graph differences (i.e., neighbor-based, path-based, and structure-based). Built upon this, we further propose DiffLens, a novel visualization system to help graph sampling developers and users intuitively explore local differences at different regions of their interest within a sampled graph, where three new lens-based visual designs are presented to display the neighbor-based, path-based, and structure-based differences respectively. We conducted two case studies and a user study using real-world network datasets to evaluate DiffLens. The results confirmed its effectiveness and usability in helping users explore local differences and compare different graph sampling strategies.

cs.HC↗

arXiv2026

Uncertainty-Aware Clarification in LLM Agents with Information Gain

Large Language Model (LLM) agents often operate under underspecified user instructions, where latent uncertainty over user intent leads to erroneous tool actions. To address this challenge, we propose a goal-oriented clarification framework that aligns clarification behavior with ambiguity resolution. Central to our approach is the Information Gain Reward, a metric that quantifies the utility of clarification questions by measuring the Bayesian belief update towards the ground-truth goal induced by the clarification exchange. We train the clarifier (LLM) using this reward to optimize for high information gain, ensuring that clarifications effectively reduce uncertainty and improve task completion within the agent-tool-user environment. We validate our framework within a clarification-enhanced $\tau$-Bench environment, conducting cross-agent evaluations across five heterogeneous backbones. Empirical results demonstrate that our method consistently improves the success rate by 3.7\% over the no-clarification baseline, while adding only 0.3 total interaction steps on average.

cs.AI↗

arXiv2026

HiSE: A Lightweight Hierarchical Semantic Explainer for Heterogeneous Graph Neural Networks

Heterogeneous graph neural networks (HGNNs) have demonstrated remarkable performance in modeling complex relational data, however their interpretability in high-stakes applications remains a critical challenge. Existing explanation methods suffer from two major limitations: on the one hand, the generated explanations fail to reflect the inherent semantic hierarchy of HGNNs, resulting in a lack of fidelity to the model's internal decision-making mechanism; on the other hand, feature explanations often rely on complex search or perturbation mechanisms, leading to excessive computational complexity and poor efficiency. To address these issues, we propose HiSE, a lightweight feature-oriented interpretable model for HGNNs. HiSE achieves semantically aware feature explanations through hierarchical semantic modeling: at the semantic level, local surrogate models based on the Least Absolute Shrinkage and Selection Operator (LASSO) are employed to learn sparse feature representations under each semantic view; at the cross-semantic level, the contributions of different semantic views are adaptively characterized via KL divergence to produce a unified explanation. Extensive experiments demonstrate that HiSE outperforms existing methods in terms of fidelity, robustness, and cross-semantic explanation capability, while its lightweight framework incurs low computational overhead, enabling efficient application to large-scale, complex real-world heterogeneous graphs.

cs.LG↗

arXiv2026

Kill Webs by Collaborative & Self-organizing Agents (CSOAs)

A single agent represents a single system capable of ingesting local data, indexing, cataloging information, performing knowledge pattern discovery, and separating patterns and anomalies from data. Multiple agents work collaboratively in a peer-to-peer network. Each agent has a peer list. Such multiple agents' collaboration can be modeled as cooperative games. Each agent optimizes its own objective locally. We show that each agent self-organizes or converges to its best value and the whole agent network achieves the best social welfare based on both the quantum adiabatic evolution transformation (QAET), and quantum intelligence game (QIG) or the QAET-QIG framework. We apply the QAET-QIG framework to the kill web concept that can potentially improve the traditional kill chain process or the find, fix, track, target, engage, and assess (F2T2EA) process. The improvement is measured in the values of powerful global optimization, distributed lethality, and load balancing. We show a use case of the QAET-QIG frame in a potential application of mixed sensors, platforms, weapons, and effects.

cs.ET↗

arXiv2026

"It from Bit": The Hartle-Hawking state and quantum mechanics for de Sitter observers

The one-state statement for closed universes has sparked considerable discussion. In this paper, we examine its physical meaning in the context of the Hartle-Hawking state and de Sitter space. We argue that the one-state property of closed universes is fully compatible with the finite-dimensional quantum mechanics experienced by observers inside de Sitter space, and that this compatibility requires neither mixing of $\alpha$-sectors nor any modification of the rules of the gravitational path integral. The apparent tension is resolved by sharply distinguishing the baby-universe Hilbert space, namely the space of closed universes viewed from the outside, from the bulk Hilbert space that governs quantum mechanics for an observer inside a single de Sitter universe. The baby-universe Hilbert space, together with its commutative operator algebra, is not a quantum-mechanical Hilbert space: it is merely a mathematical repackaging of classical probability theory and carries no quantum-mechanical structure at all, a direct consequence of the one-state property of closed universes. Accordingly, attempting to formulate quantum mechanics directly on the baby-universe Hilbert space conflates two logically distinct structures and leads to physically incorrect conclusions. By contrast, the quantum mechanics experienced by an observer inside de Sitter space emerges from the classical statistics encoded in the baby-universe Hilbert space, providing a concrete realization of Wheeler's idea of "It from Bit." We demonstrate these features by completely solving a topological toy model of one-dimensional de Sitter spacetime. Along the way, we clarify the physical meaning of de Sitter entropy, showing that it corresponds to the coarse-grained entropy of the underlying state.

hep-th↗

arXiv2025

Revealing the Partially Coherent Nature of Transport in IGZO

Thin-film transistors based on amorphous oxide semiconductors (AOS) are promising candidates for enabling further DRAM scaling and 3D integration, which are critical for advanced computing. Despite extensive research, the charge transport mechanism in these disordered semiconductors remains poorly understood. In this work, we investigate charge transport in the archetypical AOS material, indium gallium zinc oxide (IGZO), across a range of compositions and temperatures using thin-film transistors and Hall bar structures. Our results show that the electrons involved in transport exhibit partially spatial coherence and non-degenerate conduction. Under these conditions, transport is dominated by electron transfer across insulating gaps between locally coherent regions, rather than by degenerate percolative transport above a mobility edge, or by localized-state hopping, both of which are widely assumed in the literature. While fluctuation-induced tunnelling has previously been invoked to describe low-temperature transport in oxide transistors, we show that such behavior originates from partially coherent electronic states and develop a field-effect-aware fluctuation-induced tunnelling (FEAFIT) framework that explicitly accounts for gate modulation of the tunneling landscape. The FEAFIT model accurately predicts experimental data across all compositions, temperatures, and gate voltages, enabling extraction of fundamental transport parameters. These tunnelling parameters are then correlated with electron coherence dimensions and the degree of energetic disorder obtained from first-principles calculations. Our findings advance the fundamental understanding of charge transport in AOS-based transistors and provide a foundation for further performance improvements

cond-mat.mtrl-sci↗

arXiv2025

Holographic Black Hole Formation and Scrambling in Time-Ordered Correlators

We describe a holographic mechanism for black hole formation via the collision of two shock waves in three-dimensional anti-de Sitter spacetime. In the dual conformal field theory (CFT), a two-shock-wave state corresponds to the insertion of two boosted precursor operators in complementary Rindler patches. Their operator product expansion is initially described by a universal mean-field spectrum of exchanged states, which is dominated by operator dimensions that grow exponentially in the boost parameter. We propose their mean value as diagnosing the mass of the collision product in the bulk. It crosses the CFT heavy state threshold after two scrambling times, in accordance with expectations about black hole formation in general relativity. Our analysis also allows us to identify the scrambling characteristics usually associated with out-of-time-order correlation functions, using only the internal composition of thermal in-time-order correlators.

hep-th↗

arXiv2025

PCLVis: Visual Analytics of Process Communication Latency in Large-Scale Simulation

Large-scale simulations on supercomputers have become important tools for users. However, their scalability remains a problem due to the huge communication cost among parallel processes. Most of the existing communication latency analysis methods rely on the physical link layer information, which is only available to administrators. In this paper, a framework called PCLVis is proposed to help general users analyze process communication latency (PCL) events. Instead of the physical link layer information, the PCLVis uses the MPI process communication data for the analysis. First, a spatial PCL event locating method is developed. All processes with high correlation are classified into a single cluster by constructing a process-correlation tree. Second, the propagation path of PCL events is analyzed by constructing a communication-dependency-based directed acyclic graph (DAG), which can help users interactively explore a PCL event from the temporal evolution of a located PCL event cluster. In this graph, a sliding window algorithm is designed to generate the PCL events abstraction. Meanwhile, a new glyph called the communication state glyph (CS-Glyph) is designed for each process to show its communication states, including its in/out messages and load balance. Each leaf node can be further unfolded to view additional information. Third, a PCL event attribution strategy is formulated to help users optimize their simulations. The effectiveness of the PCLVis framework is demonstrated by analyzing the PCL events of several simulations running on the TH-1A supercomputer. By using the proposed framework, users can greatly improve the efficiency of their simulations.

cs.CV↗

arXiv2025

Subthreshold Swing Behavior in Amorphous Indium-Gallium-Zinc-Oxide Transistors from Room to Cryogenic Temperatures

While cryogenic-temperature subthreshold swing (SS) in crystalline semiconductors has been widely studied, a careful study on the temperature-dependent SS in amorphous oxide semiconductors remains lacking. In this paper, a comprehensive analysis of the SS in thin-film transistors with an amorphous indium gallium zinc oxide (IGZO) channel at temperatures from 300 K down to 4 K is presented. Main observations include: 1) At room temperature (300 K), the devices exhibit a SS of 61 mV/dec, and a low interface trap density (<1011 cm-2), among the best reported values for IGZO devices. 2) A SS saturation around 40 mV/dec is observed between 200 K and 100 K. It is well explained by the electron transport via band tail states with exponential decay (Wt) of 13 meV. 3) At deep-cryogenic temperature, SS increase significantly exceeding 200 mV/dec at 4 K. Such high SS values are actually limited by the measurement current range, confirmed by Id-Vg simulations based on the variable range hopping (VRH) model. This work not only elucidates the SS behavior in amorphous IGZO devices but also provides a deep understanding of the physical mechanisms of electron transport in amorphous semiconductors.

cond-mat.mtrl-sci↗

arXiv2025

Electronic Correlations Control Interlayer Coupling and Magnetic Transition in MnBi$_2$Te$_4$/MnBr$_3$ Heterostructure

Bulk MnBi$_2$Te$_4$ (MBT) is an intrinsic antiferromagnetic topological insulator. However, its low N\'eel temperature of $\sim 25\,\mathrm{K}$ severely restricts its practical applications. Here, we propose a van der Waals heterostructure composed of monolayer MBT (ML-MBT) and monolayer MnBr$_3$, an intrinsic Chern insulator possessing a high Curie temperature ($T_\mathrm{C} \sim 200\,\mathrm{K}$). By employing density functional theory calculations and Monte Carlo simulations, we demonstrate that interfacing ML-MBT with MnBr$_3$ significantly enhances the $T_\mathrm{C}$ of ML-MBT by a factor of four to five. Electronic correlations characterized by the Hubbard parameter $U_2$ for Mn-$d$ orbitals in MnBr$_3$ play a crucial role in governing magnetic coupling within the system. At a moderate correlation strength of $U_2 = 3.0\,\mathrm{eV}$, slight structural distortions in MnBr$_3$ break intralayer symmetry, enabling robust interlayer ferromagnetic coupling and yielding a single, unified magnetic transition. Increasing $U_2$ reduces these structural distortions, weakens interlayer coupling, and induces two distinct magnetic transitions, indicating interlayer magnetic decoupling. Thus, the MBT/MnBr$_3$ heterostructure offers a novel approach for controlling magnetic order and enhancing the performance of spintronic devices.

cond-mat.mtrl-sci↗

arXiv2025

On Path to Multimodal Historical Reasoning: HistBench and HistAgent

Recent advances in large language models (LLMs) have led to remarkable progress across domains, yet their capabilities in the humanities, particularly history, remain underexplored. Historical reasoning poses unique challenges for AI, involving multimodal source interpretation, temporal inference, and cross-linguistic analysis. While general-purpose agents perform well on many existing benchmarks, they lack the domain-specific expertise required to engage with historical materials and questions. To address this gap, we introduce HistBench, a new benchmark of 414 high-quality questions designed to evaluate AI's capacity for historical reasoning and authored by more than 40 expert contributors. The tasks span a wide range of historical problems-from factual retrieval based on primary sources to interpretive analysis of manuscripts and images, to interdisciplinary challenges involving archaeology, linguistics, or cultural history. Furthermore, the benchmark dataset spans 29 ancient and modern languages and covers a wide range of historical periods and world regions. Finding the poor performance of LLMs and other agents on HistBench, we further present HistAgent, a history-specific agent equipped with carefully designed tools for OCR, translation, archival search, and image understanding in History. On HistBench, HistAgent based on GPT-4o achieves an accuracy of 27.54% pass@1 and 36.47% pass@2, significantly outperforming LLMs with online search and generalist agents, including GPT-4o (18.60%), DeepSeek-R1(14.49%) and Open Deep Research-smolagents(20.29% pass@1 and 25.12% pass@2). These results highlight the limitations of existing LLMs and generalist agents and demonstrate the advantages of HistAgent for historical reasoning.

cs.AI↗

arXiv2025

Polymer Data Challenges in the AI Era: Bridging Gaps for Next-Generation Energy Materials

The pursuit of advanced polymers for energy technologies, spanning photovoltaics, solid-state batteries, and hydrogen storage, is hindered by fragmented data ecosystems that fail to capture the hierarchical complexity of these materials. Polymer science lacks interoperable databases, forcing reliance on disconnected literature and legacy records riddled with unstructured formats and irreproducible testing protocols. This fragmentation stifles machine learning (ML) applications and delays the discovery of materials critical for global decarbonization. Three systemic barriers compound the challenge. First, academic-industrial data silos restrict access to proprietary industrial datasets, while academic publications often omit critical synthesis details. Second, inconsistent testing methods undermine cross-study comparability. Third, incomplete metadata in existing databases limits their utility for training reliable ML models. Emerging solutions address these gaps through technological and collaborative innovation. Natural language processing (NLP) tools extract structured polymer data from decades of literature, while high-throughput robotic platforms generate self-consistent datasets via autonomous experimentation. Central to these advances is the adoption of FAIR (Findable, Accessible, Interoperable, Reusable) principles, adapted to polymer-specific ontologies, ensuring machine-readability and reproducibility. Future breakthroughs hinge on cultural shifts toward open science, accelerated by decentralized data markets and autonomous laboratories that merge robotic experimentation with real-time ML validation. By addressing data fragmentation through technological innovation, collaborative governance, and ethical stewardship, the polymer community can transform bottlenecks into accelerants.

cond-mat.soft↗