arXiv subjects

Han Nguyen

Publications and source records attributed to Han Nguyen.

7 recordsLinked to original sources

arXiv2026

When Agents Talk: Discourse, Manipulation, and Risk in an Agentic Social Network

AI agents are increasingly interacting within shared online environments, creating new operational security risks. We analyze activity on Moltbook, a Reddit-style social platform where AI agents--typically configured and overseen by human operators--post and interact with one another at scale. Using a dataset of 228,684 posts produced by more than 39,500 accounts over a seventeen-day observation window, we combine semantic clustering of high-engagement posts with LLM-assisted classification of harmful content and manual review of high-risk samples. The analysis identifies 98 thematic discourse clusters spanning agent infrastructure, autonomy debates, and financial activity. While most observed content was benign, 18.28% of posts contained toxic, manipulative, or malicious material. We cluster malicious content and identify 74 classes of malicious behavior, including credential harvesting attempts, host-execution instructions, proxy routing guidance, and efforts to install untrusted agent skills. Harmful content frequently appeared within mainstream operational discussions about agent functionality. We also document coordinated posting campaigns capable of generating thousands of posts in minutes.

cs.SI↗

arXiv2025

From Spheres to Cones: Structural Instabilities and Acidity at Conical Regions in Trivalent Metal Ion Nano-clusters

Sub-nanometer aqueous clusters containing a single trivalent metal cation can exhibit charge-induced structural instabilities. Here, we present computational evidence that clusters containing a single \ce{Fe^{3+}}, \ce{Lu^{3+}}, or \ce{La^{3+}} ion undergo continuous geometric transformations as a consequence of this instability. These clusters dynamically evolve across their potential energy landscape, adopting triangular, elongated two-point, single-point, and more spherical configurations often with distinct conical surface protrusions. The manifestation of this instability differs from that observed in mesoscopic and microscopic droplets containing macroions, where stable ``star-like'' structures form, characterized by a specific number of conical protrusions that varies with the droplet size. In the present study, we find that the orientation of the \ce{H2O} molecules surrounding the metal ion is influenced not only by the electric field of the trivalent ion but also by the local conical protrusions. To further investigate the local acidity in the conical protrusions, we employ a proxy model system consisting of an aqueous nano-cluster containing three \ce{H3O+} ions, simulated using ab initio molecular dynamics. Within the conical regions of the cluster, protons exhibit diffusion across several water molecules, in contrast to the more localized proton delocalization observed in the compact body of the cluster. These findings suggest that local geometry can significantly modulate acidity in highly charged nano-clusters, with potential implications for understanding charge-transfer and ionization mechanisms in techniques such as electrospray ionization mass spectrometry and complement interpretations from infrared spectroscopic data.

cond-mat.soft↗

arXiv2022

On the Algorithmic Stability and Generalization of Adaptive Optimization Methods

Despite their popularity in deep learning and machine learning in general, the theoretical properties of adaptive optimizers such as Adagrad, RMSProp, Adam or AdamW are not yet fully understood. In this paper, we develop a novel framework to study the stability and generalization of these optimization methods. Based on this framework, we show provable guarantees about such properties that depend heavily on a single parameter $\beta_2$. Our empirical experiments support our claims and provide practical insights into the stability and generalization properties of adaptive optimization methods.

cs.LG↗

arXiv2022

Freestanding LiPON: from Fundamental Study to Uniformly Dense Li Metal Deposition Under Zero External Pressure

Lithium phosphorus oxynitride (LiPON) is a well-known amorphous thin film solid electrolyte that has been extensively studied in the last three decades. Despite the promises to pair with Li metal anode and various cathode materials, the presence of rigid substrate and LiPONs unique amorphous, air-sensitive nature set limitations to comprehensively understand its intrinsic properties for future development and applications. This work demonstrates a methodology to synthesize LiPON in a freestanding form that exhibits remarkable flexibility and a Young s modulus of ~33 GPa. Solid-state nuclear magnetic resonance (ss-NMR) and differential scanning calorimetry (DSC) results with unprecedented high signal-to-noise ratio could be obtained with such freestanding LiPON (FS-LiPON), revealing the Li-LiPON interface bonding environments quantitatively and a well-defined glass transition temperature for LiPON. Combining interfacial stress and a seeding layer, FS-LiPON demonstrates a uniform and fully dense Li metal deposition without the aid of external pressure. Such a FS-LiPON film offers new opportunities for fundamental study of LiPON material and associated interfaces, and provides perspectives for interface engineering in bulk solid-state battery.

cond-mat.mtrl-sci↗

arXiv2020

Unveiling the Stable Nature of the Solid Electrolyte Interphase between Lithium Metal and LiPON via Cryogenic Electron Microscopy

The solid electrolyte interphase (SEI) is regarded as the most complex but the least understood constituent in secondary batteries using liquid and solid electrolytes. The nanostructures of SEIs were recently reported to be equally important to the chemistry of SEIs for stabilizing Li metal in liquid electrolyte. However, the dearth of such knowledge in all-solid-state battery (ASSB) has hindered a complete understanding of how certain solid-state electrolytes, such as LiPON, manifest exemplary stability against Li metal. Characterizing such solid-solid interfaces is difficult due to the buried, highly reactive, and beam-sensitive nature of the constituents within. By employing cryogenic electron microscopy (cryo-EM), the interphase between Li metal and LiPON is successfully preserved and probed, revealing a multilayer mosaic SEI structure with concentration gradients of nitrogen and phosphorous, materializing as crystallites within an amorphous matrix. This unique SEI nanostructure is less than 80 nm and is shown stable and free of any organic lithium containing species or lithium fluoride components, in contrast to SEIs often found in state-of-the-art organic liquid electrolytes. Our findings reveal insights on the nanostructures and chemistry of such SEIs as a key component in lithium metal batteries to stabilize Li metal anode.

cond-mat.mtrl-sci↗

arXiv2020

Adaptive Sampling Distributed Stochastic Variance Reduced Gradient for Heterogeneous Distributed Datasets

We study distributed optimization algorithms for minimizing the average of \emph{heterogeneous} functions distributed across several machines with a focus on communication efficiency. In such settings, naively using the classical stochastic gradient descent (SGD) or its variants (e.g., SVRG) with a uniform sampling of machines typically yields poor performance. It often leads to the dependence of convergence rate on maximum Lipschitz constant of gradients across the devices. In this paper, we propose a novel \emph{adaptive} sampling of machines specially catered to these settings. Our method relies on an adaptive estimate of local Lipschitz constants base on the information of past gradients. We show that the new way improves the dependence of convergence rate from maximum Lipschitz constant to \emph{average} Lipschitz constant across machines, thereby, significantly accelerating the convergence. Our experiments demonstrate that our method indeed speeds up the convergence of the standard SVRG algorithm in heterogeneous environments.

cs.LG↗

arXiv2019

Stack Pressure Considerations for Room Temperature All-Solid-State Lithium Metal Batteries

All-solid-state batteries are expected to enable batteries with high energy density with the use of lithium metal anodes. Although solid electrolytes are believed to be mechanically strong enough to prevent lithium dendrites from propagating, various reports today still show cell failure due to lithium dendritic growth at room temperature. While cell parameters such as current density, electrolyte porosity and interfacial properties have been investigated, mechanical properties of lithium metal and the role of applied stack pressure on the shorting behavior is still poorly understood. Here, we investigated failure mechanisms of lithium metal in all-solid-state batteries as a function of stack pressure, and conducted in situ characterization of the interfacial and morphological properties of the buried lithium in solid electrolytes. We found that a low stack pressure of 5 MPa allows reliable plating and stripping in a lithium symmetric cell for more than 1000 hours, and a Li | Li6PS5Cl | LiNi0.80Co0.15Al0.05O2 full cell, plating more than 4 um of lithium per charge, is able to cycle over 200 cycles at room temperature. These results suggest the possibility of enabling the lithium metal anode in all-solid-state batteries at reasonable stack pressures.

physics.app-ph↗