arXiv subjects

Daniel Escudero

Publications and source records attributed to Daniel Escudero.

11 recordsLinked to original sources

arXiv2026

Scalable Secure Biometric Authentication without Auxiliary Identifiers

The prevalence of biometric authentication has been on the rise due to its ease of use and elimination of weak passwords. To date, most biometric authentication systems have been designed for on-device authentication of the device owner (e.g., smartphones and laptops). Recently, biometric authentication systems have started to emerge that are designed to authenticate users against cloud databases storing representations of biometrics for large numbers of users (potentially millions), such as those facilitating biometric payments. However, the use of a large cloud database introduces a significant attack vector, as a breach of the database could lead to the compromise of all enrolled users' sensitive biometric data. Indeed, all such existing systems either do not adequately protect against such a breach, or are impractical to deploy and use due to their high computational overhead. In this work, we present a new biometric authentication system that provides provable security guarantees against data breaches, while remaining scalable and performant. To do so, we marry artificial intelligence with advanced cryptographic techniques in a novel fashion, providing several optimizations along the way. Our work is the first to show that real-world scalable privacy-preserving biometric authentication without auxiliary identifiers is feasible, and we believe that it will spur widespread industrial adoption and further research in this area.

cs.CR↗

arXiv2025

Quantifying the Role of Higher-Lying Excited States in Organic Emitters via Multistate Ab Initio Kinetic Modeling

Higher lying excited states beyond S1 and T1 are widely recognized in many photophysical systems, including thermally activated delayed fluorescence (TADF). However, their explicit and quantitative impact on photophysical observables such as photoluminescence quantum yields (PLQY) and lifetimes is difficult to be attained experimentally and it has not been systematically assessed within a fully ab initio kinetic modeling framework. To address this gap, we developed KinLuv, a multistate excited state kinetic model that includes higher lying excited states (S2, T2) and all possible monomolecular interconversion processes between all the electronic states, whose rate constants were computed using Fermi golden rule explicitly including the Herzberg Teller (HT) vibronic coupling effect. We applied KinLuv to prototypical multi resonance TADF (MRTADF) emitters and their derivatives, as well as other representative organic chromophores, demonstrating its broad applicability across diverse photophysical playgrounds beyond TADF. The resulting simulations quantitatively reproduce key experimental observables, including PLQY and prompt and delayed fluorescence lifetimes. Beyond its predictive power, the present results establish clear criteria for identifying when higher lying excited states influence the excited state decay and when simplified models remain adequate. This framework enables rational selection of minimal kinetic models that balance physical insight with numerical robustness, with direct implications for the in silico design of high performance organic emitters.

physics.chem-ph↗

arXiv2025

Covert Attacks on Machine Learning Training in Passively Secure MPC

Secure multiparty computation (MPC) allows data owners to train machine learning models on combined data while keeping the underlying training data private. The MPC threat model either considers an adversary who passively corrupts some parties without affecting their overall behavior, or an adversary who actively modifies the behavior of corrupt parties. It has been argued that in some settings, active security is not a major concern, partly because of the potential risk of reputation loss if a party is detected cheating. In this work we show explicit, simple, and effective attacks that an active adversary can run on existing passively secure MPC training protocols, while keeping essentially zero risk of the attack being detected. The attacks we show can compromise both the integrity and privacy of the model, including attacks reconstructing exact training data. Our results challenge the belief that a threat model that does not include malicious behavior by the involved parties may be reasonable in the context of PPML, motivating the use of actively secure protocols for training.

cs.CR↗

arXiv2025

Peak Broadening in Photoelectron Spectroscopy of Amorphous Polymers: the Leading Role of the Electrostatic Landscape

The broadening in photoelectron spectra of polymers can be attributed to several factors, such as light source spread, spectrometer resolution, finite lifetime of the hole state, and solid-state effects. Here, for the first time, we set up a computational protocol to assess the peak broadening induced for both core and valence levels by solid-state effects in four amorphous polymers by using a combination of density functional theory, many-body perturbation theory, and classical polarizable embedding. We show that intrinsic local inhomogeneities in the electrostatic environment induce a Gaussian broadening of $0.2$-$0.7$~eV in the binding energies of both core and semi-valence electrons, corresponding to a full width at half maximum (FWHM) of $0.5$-$1.7$~eV for the investigated systems. The induced broadening is larger in acrylate- than in styrene- based polymers, revealing the crucial role of polar groups in controlling the roughness of the electrostatic landscape in the solid matrix.

cond-mat.mtrl-sci↗

arXiv2024

Electron Spin Dynamics of the Intersystem Crossing in Aminoanthraquinone Derivatives: The Spectral Telltale of Short Triplet Excited States

Herein we studied the excited state dynamics of two bis-amino substituted anthraquinone (AQ) derivatives. Femtosecond transient absorption spectra show that intersystem crossing (ISC) takes place in 190-320 ps, and nanosecond transient absorption spectra demonstrated unusually short triplet state lifetime (2.1-5.4 us) for the two AQ derivatives at room temperature. Pulsed laser excited time-resolved electron paramagnetic resonance (TREPR) spectra shows an inversion of the electron spin polarization (ESP) phase pattern of the triplet state at longer delay time. Spectral simulations show that the faster decay of the Ty sublevel (x = 15.0 us, y = 1.50 us, z = 15.0 us) rationalizes the short T1 state lifetime and the ESP inversion. Computations taking into account the electron-vibrational coupling, i.e., the Herzberg-Teller effect, successfully rationalize the TREPR experimental observations.

physics.chem-ph↗

arXiv2022

Modeling X-ray Photoelectron Spectroscopy of Macromolecules Using GW

We propose a simple additive approach to simulate X-ray photoelectron spectra (XPS) of macromolecules based on the $GW$ method. Single-shot $GW$ ($G_0W_0$) is a promising technique to compute accurate core-electron binding energies (BEs). However, its application to large molecules is still unfeasible. To circumvent the computational cost of $G_0W_0$, we break the macromolecule into tractable building blocks, such as isolated monomers, and sum up the theoretical spectra of each component, weighted by their molar ratio. In this work, we provide a first proof of concept by applying the method to four test polymers and one copolymer, and show that it leads to an excellent agreement with experiments. The method could be used to retrieve the composition of unknown materials and study chemical reactions, by comparing the simulated spectra with experimental ones.

cond-mat.mtrl-sci↗

arXiv2021

Secure Evaluation of Quantized Neural Networks

We investigate two questions in this paper: First, we ask to what extent "MPC friendly" models are already supported by major Machine Learning frameworks such as TensorFlow or PyTorch. Prior works provide protocols that only work on fixed-point integers and specialized activation functions, two aspects that are not supported by popular Machine Learning frameworks, and the need for these specialized model representations means that it is hard, and often impossible, to use e.g., TensorFlow to design, train and test models that later have to be evaluated securely. Second, we ask to what extent the functionality for evaluating Neural Networks already exists in general-purpose MPC frameworks. These frameworks have received more scrutiny, are better documented and supported on more platforms. Furthermore, they are typically flexible in terms of the threat model they support. In contrast, most secure evaluation protocols in the literature are targeted to a specific threat model and their implementations are only a "proof-of-concept", making it very hard for their adoption in practice. We answer both of the above questions in a positive way: We observe that the quantization techniques supported by both TensorFlow, PyTorch and MXNet can provide models in a representation that can be evaluated securely; and moreover, that this evaluation can be performed by a general purpose MPC framework. We perform extensive benchmarks to understand the exact trade-offs between different corruption models, network sizes and efficiency. These experiments provide an interesting insight into cost between active and passive security, as well as honest and dishonest majority. Our work shows then that the separating line between existing ML frameworks and existing MPC protocols may be narrower than implicitly suggested by previous works.

cs.CR↗

arXiv2019

Facial and Meridional Isomers of Tris(bidentate) Ir(III) Complexes: Unravelling Their Different Excited State Reactivity

The use of tris(bidentate) Ir(III) complexes as light active components in phosphorescent organic light-emitting diodes (PhOLEDs) is currently the state-of-the-art technology to attain long-lasting and highly performing devices. Still, further improvements of their operational lifetimes are required for their practical use in lighting and displays. Facial/meridional stereoisomerism of the tris(bidentate) Ir(III) architectures strongly influences their emissive properties and thereto their PhOLEDs performances and operational device stabilities. This work underpins at the first-principles level the different excited state reactivities of facial and meridional stereoisomers of a series of tris(bidentate) Ir(III) complexes, which is found to originate in the presence of distinct triplet metal-centered (3MC) deactivation pathways. These deactivation pathways are herein presented for the first time for the meridional isomers. Finally, we propose some phosphor design strategies.

physics.chem-ph↗

arXiv2019

General Approach To Compute Phosphorescent OLED Efficiency

Phosphorescent organic light-emitting diodes (PhOLEDs) are widely used in the display industry. In PhOLEDs, cyclometalated Ir(III) complexes are the most widespread triplet emitter dopants to attain red, e.g., Ir(piq)3 (piq = 1-phenylisoquinoline), and green, e.g., Ir(ppy)3 (ppy = 2-phenylpyridine), emissions, whereas obtaining operative deep-blue emitters is still one of the major challenges. When designing new emitters, two main characteristics besides colors should be targeted: high photostability and large photoluminescence efficiencies. To date, these are very often optimized experimentally in a trial-and-error manner. Instead, accurate predictive tools would be highly desirable. In this contribution, we present a general approach for computing the photoluminescence lifetimes and efficiencies of Ir(III) complexes by considering all possible competing excited-state deactivation processes and importantly explicitly including the strongly temperature-dependent ones. This approach is based on the combination of state-of-the-art quantum chemical calculations and excited-state decay rate formalism with kinetic modeling, which is shown to be an efficient and reliable approach for a broad palette of Ir(III) complexes, i.e., from yellow/orange to deep-blue emitters.

physics.chem-ph↗

arXiv2019

Modelling the photochrome-TiO2 interface with Bethe-Salpeter and TD-DFT methods

Hybrid organic/inorganic-semiconductor systems have important applications in both molecular electronics and in photo-responsive materials. The characterization of the interface and of the electronic excited-states of these hybrid systems remains a challenge for state-of-the-art computational methods, as the systems of interest are large. In the present investigation, we present for the first time a many-body Green's function Bethe-Salpeter investigation of a series of photochromic molecules adsorbed onto TiO2 nanoclusters. Based on these studies, the performance of TD-DFT is assessed. Using a state-of-the-art computational protocol, the photochromic properties of different hybrid systems are assessed. This work shows that qualitatively different conclusions can be reached with TD-DFT relying on various exchange-correlation functionals for such organic/inorganic interfaces, and paves the way to more accurate simulation of many materials.

physics.chem-ph↗

arXiv2019

Thermal equilibration between excited states or solvent effects: unveiling the origins of anomalous emissions in heteroleptic Ru(II) complexes

In this manuscript we present a computational study on the photoluminescence properties of several heteroleptic [Ru(H)(CO)(N^N)(tpp)2]+ complexes (tpp = triphenylphosphine). A special focus is set on disentangling the temperature-dependent emissive properties. Experimentally, when cooling a solution of [Ru(H)(CO)(dmphen)(tpp)2]+ (dmphen = 5,6-dimethyl-1,10-phenanthroline) from room to cryogenic temperature, a partial emission switch from metal-to-ligand charge transfer (3MLCT) to ligand-centered (3LC) phosphorescence is observed, resulting in dual photoluminescence. Two different origins of the anomalous emissions are possible, i.e., thermal equilibration between electronically excited states or different excited state solvent relaxation effects. Our calculations are in favor of the thermally equilibrated scenario. This computational investigation highlights the importance of controlling the temperature-dependent emissive behavior for optoelectronic applications.

physics.chem-ph↗