arXiv subjects

Adi Levi

Publications and source records attributed to Adi Levi.

5 recordsLinked to original sources

arXiv2026

Spectrally Programmable Spin-Polarized Photocurrents in WSe$_2$-NiPS$_3$ Magnetic van der Waals Heterostructures

Efficient generation and control of spin-polarized currents in semiconductors remain central challenges for spin-based electronics, particularly due to impedance mismatch and the reliance on magnetic fields or ferromagnetic contacts. Here, we introduce a materials platform for spectrally programmable spin transport based on a van der Waals heterostructure combining the antiferromagnetic semiconductor NiPS$_3$ with WSe$_2$. In a p-n diode architecture, circularly polarized excitation produces pronounced photoconductive resonances with spin polarization reaching 80% near the Neel temperature and persisting at 30% at room temperature. Remarkably, selected spectral bands retain their polarization sign across the magnetic phase transition, evidencing robust, spectrally protected spin-polarized current generation. Polarization-resolved photogalvanic measurements reveal a dominant circular injection-current mechanism, confirming spin-polarized carrier transport. First-principles calculations show that an applied electric field induces interfacial hybridization and spin-layer locking, giving rise to localized symmetry breaking and enhanced optical absorption while preserving global time-reversal symmetry. These results establish spectral tuning of excitation as a new control knob for spin transport, enabling spin-current generation without magnetic fields or polarization switching. Our findings position magnetic van der Waals heterostructures as a versatile platform for opto-spintronic functionality and spectrally programmable spintronic devices.

cond-mat.mtrl-sci↗

arXiv2026

LLM Performance Predictors: Learning When to Escalate in Hybrid Human-AI Moderation Systems

As LLMs are increasingly integrated into human-in-the-loop content moderation systems, a central challenge is deciding when their outputs can be trusted versus when escalation for human review is preferable. We propose a novel framework for supervised LLM uncertainty quantification, learning a dedicated meta-model based on LLM Performance Predictors (LPPs) derived from LLM outputs: log-probabilities, entropy, and novel uncertainty attribution indicators. We demonstrate that our method enables cost-aware selective classification in real-world human-AI workflows: escalating high-risk cases while automating the rest. Experiments across state-of-the-art LLMs, including both off-the-shelf (Gemini, GPT) and open-source (Llama, Qwen), on multimodal and multilingual moderation tasks, show significant improvements over existing uncertainty estimators in accuracy-cost trade-offs. Beyond uncertainty estimation, the LPPs enhance explainability by providing new insights into failure conditions (e.g., ambiguous content vs. under-specified policy). This work establishes a principled framework for uncertainty-aware, scalable, and responsible human-AI moderation workflows.

cs.AI↗

arXiv2025

AI vs. Human Moderators: A Comparative Evaluation of Multimodal LLMs in Content Moderation for Brand Safety

As the volume of video content online grows exponentially, the demand for moderation of unsafe videos has surpassed human capabilities, posing both operational and mental health challenges. While recent studies demonstrated the merits of Multimodal Large Language Models (MLLMs) in various video understanding tasks, their application to multimodal content moderation, a domain that requires nuanced understanding of both visual and textual cues, remains relatively underexplored. In this work, we benchmark the capabilities of MLLMs in brand safety classification, a critical subset of content moderation for safe-guarding advertising integrity. To this end, we introduce a novel, multimodal and multilingual dataset, meticulously labeled by professional reviewers in a multitude of risk categories. Through a detailed comparative analysis, we demonstrate the effectiveness of MLLMs such as Gemini, GPT, and Llama in multimodal brand safety, and evaluate their accuracy and cost efficiency compared to professional human reviewers. Furthermore, we present an in-depth discussion shedding light on limitations of MLLMs and failure cases. We are releasing our dataset alongside this paper to facilitate future research on effective and responsible brand safety and content moderation.

cs.CV↗

arXiv2023

Nonlinear Self-Calibrated Spectrometer with Single GeSe-InSe Heterojunction Device

Optical spectroscopy the measurement of electromagnetic spectra is fundamental to various scientific domains and serves as the building block of numerous technologies. Computational spectrometry is an emerging field that employs an array of photodetectors with different spectral responses or a single photodetector device with tunable spectral response, in conjunction with numerical algorithms, for spectroscopic measurements. Compact single photodetectors made from layered materials are particularly attractive, since they eliminate the need for bulky mechanical and optical components used in traditional spectrometers and can easily be engineered as heterostructures to optimize device performance. However, compact tunable photodetectors are typically nonlinear devices and this adds complexity to extracting optical spectra from the device response. Here, we report on the training of an artificial neural network (ANN) to recover the full nonlinear spectral photoresponse of a nonlinear problem of high dimensionality of a single GeSe-InSe p-n heterojunction device. We demonstrate the functionality of a calibrated spectrometer in the spectral range of 400-1100 nm, with a small device footprint of ~25X25 micrometers, and we achieve a mean reconstruction error of 0.0002 for the power-spectrum at a spectral resolution of 0.35 nm. Using our device, we demonstrate a solution to metamerism, an apparent matching of colors with different power spectral distributions, which is a fundamental problem in optical imaging.

physics.app-ph↗

arXiv2021

Amorphous Selenium Mie Resonators for Infrared Meta-Optics

Applying direct growth and deposition of optical surfaces holds great promise for the advancement of future nanophotonic technologies. Here, we report on a chemical vapor deposition (CVD) technique for depositing amorphous selenium (a-Se) spheres by desorption of selenium from Bi2Se3 and re-adsorption on the substrate. We utilize this process to grow scalable, large area Se spheres on several substrates and characterize their Mie-resonant response in the mid-infrared (MIR) spectral range. We demonstrate size-tunable Mie resonances spanning the 2-16 um spectral range, for single isolated resonators and large area ensembles, respectively. We further demonstrate strong absorption dips of up to 90% in ensembles of particles in a broad MIR range. Finally, we show that ultra-high-Q resonances arise in the case where Se Mie-resonators are coupled to low-loss epsilon-near-zero (ENZ) substrates. These findings demonstrate the enabling potential of amorphous Selenium as a versatile and tunable nanophotonic material that may open up avenues for on-chip MIR spectroscopy, chemical sensing, spectral imaging and large area metasurface fabrication.

physics.optics↗