arXiv subjects

Jincheng An

Publications and source records attributed to Jincheng An.

14 recordsLinked to original sources

arXiv2026

Combating Knowledge Corruption in Agent Systems: A Byzantine-Tolerant Secure Collaborative RAG Framework

While retrieval-augmented generation systems partially address the hallucination issues in large language models, it also introduces new vulnerabilities to knowledge corruption attacks. Adversaries exploit these vulnerabilities by poisoning documents provided by RAG system to manipulate LLM outputs. To counter this threat, we propose SecureCollaRAG, a Byzantine-tolerant collaborative RAG framework leveraging Multi-source Knowledge Validation Mechanism. Our approach enables agent system to securely verify document provenance through dynamic GNN-based credibility scoring, effectively preventing stealthy knowledge corruption attacks while preserving essential domain knowledge integrity. Through extensive evaluations and formal analysis, we demonstrate that SecureCollaRAG maintains robustness against attackers under non-IID data distributions.

cs.CR↗

arXiv2026

Skyrmion Excitations in the $ν=-1$ Quantum Hall state in Monolayer Graphene

We investigate the skyrmion excited states atop the $SU(4)$ quantum Hall ferromagnetic ground state at filling factor $ν=-1$ in monolayer graphene. The competition among short-range anisotropic interactions, the Zeeman coupling, and sublattice symmetry-breaking potential gives rise to four distinct spin-valley-ordered ground-state phases. Combining effective field theory with the Hartree--Fock approximation, we develop a variational framework that incorporates both the long-range Coulomb interaction and the symmetry-breaking terms on an equal footing. Variational minimization determines the optimal skyrmion texture, including both its spatial radial profile and internal $SU(4)$ spinor structure. We establish the phase diagram of skyrmion excitations, identify thirteen distinct skyrmion phases under different external-field conditions, and determine their spin-valley textures, energies, and characteristic sizes. The robustness of the results against different radial ansätze demonstrates the reliability of our variational approach. Our work provides a systematic framework for studying skyrmion excitations in multicomponent quantum Hall ferromagnets and can be naturally extended to more general $SU(4)$ quantum Hall systems.

cond-mat.mes-hall↗

arXiv2026

Rényi Pufferfish Privacy with Gaussian-based Priors: From Single Gaussian to Mixture Model

Rényi Pufferfish Privacy (RPP) provides a Rényi divergence-based privacy framework for correlated data, but existing $\infty$-Wasserstein mechanisms are often conservative and sacrifice data utility. We study Gaussian mechanisms for RPP under Gaussian and Gaussian-mixture priors. For single Gaussian priors, we derive the exact Rényi divergence after Gaussian perturbation, obtain a relaxed closed-form sufficient condition for $(α,ε)$-RPP, and characterize the monotonicity of the calibrated noise with respect to the privacy budget $ε$ and the Rényi order $α$. To handle more general non-Gaussian and multimodal priors, we approximate secret-conditioned outputs with Gaussian mixture models and introduce an optimal-transport-based sufficient condition for RPP. Experiments on three UCI datasets with statistical (\textsc{RAW}, \textsc{MEAN}) and model-output (\textsc{BNN}, \textsc{GP}) queries show that our prior-aware mechanisms consistently require less noise than a recent RPP additive-noise baseline, achieving an average noise reduction of 48.9\%. These results show that our mechanisms can substantially improve the privacy-utility trade-off under RPP.

cs.CR↗

arXiv2026

Noise Reduction for Pufferfish Privacy: A Practical Noise Calibration Method

This paper introduces a relaxed noise calibration method to enhance data utility while attaining pufferfish privacy. This work builds on the existing $1$-Wasserstein (Kantorovich) mechanism by alleviating the existing overly strict condition that leads to excessive noise, and proposes a practical mechanism design algorithm as a general solution. We prove that a strict noise reduction by our approach always exists compared to $1$-Wasserstein mechanism for all privacy budgets $ε$ and prior beliefs, and the noise reduction (also represents improvement on data utility) gains increase significantly for low privacy budget situations--which are commonly seen in real-world deployments. We also analyze the variation and optimality of the noise reduction with different prior distributions. Moreover, all the properties of the noise reduction still exist in the worst-case $1$-Wasserstein mechanism we introduced, when the additive noise is largest. We further show that the worst-case $1$-Wasserstein mechanism is equivalent to the $\ell_1$-sensitivity method. Experimental results on three real-world datasets demonstrate $47\%$ to $87\%$ improvement in data utility.

cs.CR↗

arXiv2026

Jailbreaking Large Language Models through Iterative Tool-Disguised Attacks via Reinforcement Learning

Large language models (LLMs) have demonstrated remarkable capabilities across diverse applications, however, they remain critically vulnerable to jailbreak attacks that elicit harmful responses violating human values and safety guidelines. Despite extensive research on defense mechanisms, existing safeguards prove insufficient against sophisticated adversarial strategies. In this work, we propose iMIST (\underline{i}nteractive \underline{M}ulti-step \underline{P}rogre\underline{s}sive \underline{T}ool-disguised Jailbreak Attack), a novel adaptive jailbreak method that synergistically exploits vulnerabilities in current defense mechanisms. iMIST disguises malicious queries as normal tool invocations to bypass content filters, while simultaneously introducing an interactive progressive optimization algorithm that dynamically escalates response harmfulness through multi-turn dialogues guided by real-time harmfulness assessment. Our experiments on widely-used models demonstrate that iMIST achieves higher attack effectiveness, while maintaining low rejection rates. These results reveal critical vulnerabilities in current LLM safety mechanisms and underscore the urgent need for more robust defense strategies.

cs.CR↗

arXiv2025

Anomalous Transport Gaps of Fractional Quantum Hall Phases in Graphene Landau Levels are Induced by Spin-Valley Entangled Ground States

We evaluate the transport gaps in the most prominent fractional quantum Hall states in the $\mathbf{n}{=}0$ and $\mathbf{n}{=}1$ Landau Levels of graphene, accounting for the Coulomb interaction, lattice-scale anisotropies, and one-body terms. We find that the fractional phases in the $\mathbf{n}{=}0$ Landau level are bond-ordered, while those in the $\mathbf{n}{=}1$ Landau level are spin-valley entangled. This resolves a long-standing experimental puzzle [Amet, $\textit{et al.}$, Nat. Comm. $\mathbf{6}$, 5838 (2015)] of the contrasting Zeeman dependence of the transport gaps in the two Landau levels. The spin-valley entangled phases host gapless Goldstone modes that can be probed via bulk thermal transport measurements. As a byproduct of our computations, we place strong constraints on the values of the microscopic anisotropic couplings such that these are consistent with all known experimental results.

cond-mat.mes-hall↗

arXiv2024

Uniquely identifying quantum Hall phases in charge neutral graphene

Charge-neutral graphene in the quantum Hall regime is an example of a quantum Hall ferromagnet in a complex spin-valley space. This system exhibits a plethora of phases, with the particular spin-valley order parameters chosen by the system depending sensitively on the short-range anisotropic couplings, the Zeeman field, and the sublattice symmetry breaking field. A subset of order parameters related to lattice symmetry-breaking have been observed by scanning tunneling microscopy. However, other order parameters, particularly those which superpose spin and valley, are more elusive, making it difficult to pin down the nature of the phase. We propose a solution this problem by examining two types of experimentally measurable quantities; transport gaps and collective mode dispersions. We find that the variation of the transport gap with the Zeeman and sublattice symmetry breaking fields, in conjunction with the number of Larmor and gapless modes, provides a unique signature for each theoretically possible phase.

cond-mat.mes-hall↗

arXiv2024

Mixed Delay/Nondelay Embeddings Based Neuromorphic Computing with Patterned Nanomagnet Arrays

Patterned nanomagnet arrays (PNAs) have been shown to exhibit a strong geometrically frustrated dipole interaction. Some PNAs have also shown emergent domain wall dynamics. Previous works have demonstrated methods to physically probe these magnetization dynamics of PNAs to realize neuromorphic reservoir systems that exhibit chaotic dynamical behavior and high-dimensional nonlinearity. These PNA reservoir systems from prior works leverage echo state properties and linear/nonlinear short-term memory of component reservoir nodes to map and preserve the dynamical information of the input time-series data into nondelay spatial embeddings. Such mappings enable these PNA reservoir systems to imitate and predict/forecast the input time series data. However, these prior PNA reservoir systems are based solely on the nondelay spatial embeddings obtained at component reservoir nodes. As a result, they require a massive number of component reservoir nodes, or a very large spatial embedding (i.e., high-dimensional spatial embedding) per reservoir node, or both, to achieve acceptable imitation and prediction accuracy. These requirements reduce the practical feasibility of such PNA reservoir systems. To address this shortcoming, we present a mixed delay/nondelay embeddings-based PNA reservoir system. Our system uses a single PNA reservoir node with the ability to obtain a mixture of delay/nondelay embeddings of the dynamical information of the time-series data applied at the input of a single PNA reservoir node. Our analysis shows that when these mixed delay/nondelay embeddings are used to train a perceptron at the output layer, our reservoir system outperforms existing PNA-based reservoir systems for the imitation of NARMA 2, NARMA 5, NARMA 7, and NARMA 10 time series data, and for the short-term and long-term prediction of the Mackey Glass time series data.

cs.ET↗

arXiv2024

Fractional quantum Hall coexistence phases in higher Landau levels of graphene

Monolayer graphene under a strong magnetic field near charge neutrality manifests the integer and fractional quantum Hall effects. Since only some of the four spin/valley flavors available to the electrons in each Landau level manifold are filled, they also exhibit spontaneous symmetry breaking the in spin/valley sector, a phenomenon known as quantum Hall ferromagnetism. In this work, we study quantum Hall ferromagnets in the higher Landau level manifolds of monolayer graphene and show that there is an even richer set of symmetry-broken phases than in the lowest Landau level manifold. Specifically, both valley polarized and valley equatorial (where the occupied Landau levels are in an equal superposition of both valleys) ferromagnets, antiferromagnets, and canted antiferromagnets are found. Several types of spin valley entangled phases are found, all of which manifest the simultaneous spontaneous symmetry breaking of both magnetic and lattice symmetries.

cond-mat.mes-hall↗

arXiv2024

Magnetic and Lattice Ordered Fractional Quantum Hall Phases in Graphene

At and near charge neutrality, monolayer graphene in a perpendicular magnetic field is a quantum Hall ferromagnet. In addition to the highly symmetric Coulomb interaction, residual lattice-scale interactions, Zeeman, and sublattice couplings determine the fate of the ground state. Going beyond the simplest model with ultra-short-range residual couplings to more generic couplings, one finds integer phases that show the coexistence of magnetic and lattice order parameters. Here we show that fractional quantum Hall states in the vicinity of charge neutrality have even richer phase diagrams, with a plethora of phases with simultaneous magnetic and lattice symmetry breaking.

cond-mat.mes-hall↗

arXiv2018

5-dimensional Myers-Perry Black Holes Cannot be Over-spun by Gedanken Experiments

We apply the new version of gedanken experiment designed recently by Sorce and Wald, to over-spin the 5-dimensional Myers-Perry black holes. As a result, the extremal black holes cannot be over-spun at the linear order. On the other hand, although the nearly extremal black holes could be over-spun at the linear order, this process is shown to be prohibited by the quadratic order correction. Thus no violation of the weak cosmic censorship conjecture occurs around the 5-dimensional Myers-Perry black holes.

hep-th↗

arXiv2017

Naked Singularity as Accelerator for Spinning Particle

When two particles collide at the horizon of a black hole and one of them satisfies some critical conditions, the relative velocity between them can be arbitrarily large, thus the energy of the center-of-mass will reach infinity. Such a process is called BSW mechanism which can accelerate a particle to arbitrarily high energy. There are also some studies showing that a Kerr naked singularity can be more qualified as a particle accelerator for arbitrarily high energy. Previous researchers mainly concentrate on geodesic motion of particles. In this paper, we will take spinning particles which won't move along a timelike geodesic and carry more parameters into our consideration. By employing the Mathisson-Papapetrou-Dixon equation, we will prove that for a spinning particle in hyper-extremal Reissner-Nordstrom or Kerr spacetime where exists a naked singularity at $r=0$, its Effective Potential $V_{eff}=-\dot{r}^2$ must be able to reach zero within the interval $0 < r < M$, thus an ingoing particle will be able to turn back and then collide with another ingoing particle at $r=M$. If the spacetime is ${\bf slightly}$ hyper-extremal, the energy of center of mass $E_{cm}$ will be arbitrarily high.

gr-qc↗

arXiv2017

Kerr-Sen Black Hole as Accelerator for Spinning Particles

It has been proved that arbitrarily high-energy collision between two particles can occur near the horizon of an extremal Kerr black hole as long as the energy $E$ and angular momentum $L$ of one particle satisfies a critical relation, which is called the BSW mechanism. Previous researchers mainly concentrate on geodesic motion of particles. In this paper, we will take spinning particle which won't move along a timelike geodesic into our consideration, hence, another parameter $s$ describing the particle's spin angular momentum was introduced. By employing the Mathisson-Papapetrou-Dixon equation describing the movement of spinning particle, we will explore whether a Kerr-Sen black hole which is slightly different from Kerr black hole can be used to accelerate a spinning particle to arbitrarily high energy. We found that when one of the two colliding particles satisfies a critical relation between the energy $E$ and the total angular momentum $J$, or has a critical spinning angular momentum $s_c$, a divergence of the center-of-mass energy $E_{cm}$ will be obtained.

gr-qc↗

arXiv2017

Five-dimensional Myers-Perry Black Holes as Particle Accelerators

It has been shown that black holes could be used as particle accelerators to create arbitrarily high center-of-mass (CM) energy if certain critical conditions are satisfied. Most studies so far are confined in four-dimensional spacetimes. In this paper, we present a systematic analysis on five-dimensional Myers-Perry black holes and find some novel properties compared to four-dimensional Kerr black holes. Firstly, we give a rigorous proof that untrhigh energy collisions cannot occur near a five-dimensional nonextremal black hole. Secondly, For extremal black holes, we find a critical condition on the particles' parameters causing ultraenergetic collisions. Thirdly, when the spacetime contains a naked singularity, we show that the CM energy could diverge at the singularity if one of the particle just bounces back at the singularity. Finally, we explore a special and important case where the naked singularity just begins to form. Surprisingly, the ultraenergetic collisions do not need any fine-turning in that case. However, we find that one of the conserved angular momentums must be nonzero.

gr-qc↗