Repository for Publications and Research Data
News from the ETH Library
Recently Added
Trap-to-trap free falls with an optically levitated nanoparticle
Item type: Journal Article
M. Luisa Mattana; Nicola Carlon Zambon; Massimiliano Rossi; et al. (2026)
We perform free-fall experiments with a charge-neutral, optically levitated nanoparticle. This is achieved using an optical tweezer that can be rapidly toggled on and off and vertically displaced, enabling the particle to be released and recaptured after each free fall. The particle is insensitive to electric fields due to its charge neutrality, and during free evolution, is not subject to photon recoil heating. We achieve free-fall durations of up to 0.25ms and observe a nearly 200-fold increase in the particle's position uncertainty at recapture. The current limit on the free-fall time arises from the performance of the initial cooling step. By implementing linear feedback techniques and reducing the background pressure, we expect to perform millisecond-scale free-fall experiments in an ultrahigh vacuum, opening new opportunities for generating large delocalizations of levitated objects.
Phishing for proficiency: evaluating proficiency-based categorisation for personalised phishing training
Item type: Journal Article
Schöni, Lorin; Roch, Neele; Carles, Victor; et al. (2026)
Purpose - This study aims to investigate the impact of personalised phishing training on users’ phishing detection skills, by adapting training content to users’ phishing proficiency. The authors provide practical recommendations on how phishing training can be improved through personalisation.
Design/methodology/approach - In two online studies with 96 and 158 participants, the authors assigned participants to one of three groups that received tailored training based on a composite phishing proficiency measure.
Findings - The training enhanced overall phishing proficiency and reduced disparities between participants, by equalising proficiency across groups, regardless of their initial proficiency. These effects transferred to phishing classification accuracy, which supports the utility of the proficiency-based grouping approach.
Originality/value - This work advances personalised phishing training by introducing a composite phishing proficiency score, revising it and empirically validating its effectiveness and demonstrating that sparse pre-training data can enable personalised and efficient training. The authors provide an empirically tested foundation for tailoring interventions by mapping users to training modules based on their proficiency, rather than static demographic or personality traits.
Partially joint petrophysical inversion
Item type: Journal Article
Söding, Hagen; Wagner, Florian M.; Maurer, Hansruedi (2026)
Joint petrophysical inversion is a powerful technique for using multiple geophysical modalities to estimate petrophysical or geotechnical parameters of the subsurface. A precise knowledge of the petrophysical laws for the full model domain is imperative to enable petrophysical coupling. In this work, we investigate the effect of partially invalid petrophysical laws on the inversion of a synthetic data set, using electrical resistivity tomography (ERT) and seismic traveltime data to image a CO₂ plume in a Carbon Capture and Storage (CCS) setup. We consider a model consisting of a reservoir and a caprock in which only the reservoir can be described by a petrophysical law. We first apply a conventional (joint) petrophysical inversion (JPI) and show that the use of wrong petrophysical laws leads to systemic artefacts within the parts of the model in which the petrophysical relations are invalid. We then present a new hybrid partially petrophysically coupled joint inversion (P-JPI) approach that combines petrophysical coupling for regions with valid petrophysical laws, and structural coupling, whenever no reliable petrophysical laws are available. The P-JPI approach outperforms tomography based on the individual ERT or seismic data set, as well as joint structural inversion (JSI) based on the cross-gradient functional. The partially petrophysically coupled joint inversion thus enables petrophysical coupling and provides a unique, quantitatively interpretable saturation model for the CO₂-plume. We further show that it is possible to detect zones with incorrect petrophysical relations by analysing the difference of the model updates based on the standalone data sets. Finally, we combine the detection of zones of incorrect petrophysical laws with the P-JPI to derive an inversion scheme that is independent of prior knowledge of the validity of petrophysical laws. Our novel methods facilitate direct estimation of the petrophysical subsurface parameters from multiple geophysical measurements if petrophysical relations are only available for parts of the model domain and provide means to quantify the spatial extent of regions where the petrophysical relations are valid.
Steerable Zero-Shot Neural Architecture Search for Efficient Edge Inference
Item type: Conference Paper
Narduzzi, Simon; Vuagniaux, Remy; Sharma, Kishan; et al. (2025)
Recent advancements in Neural Architecture Search (NAS) have introduced methods capable of identifying optimal neural network architectures in minutes on Graphical Processing Units (GPUs) using zero-shot proxies, but mainly focus on single-objective optimization. NAS is also used to discover efficient architectures for edge devices. However, addressing the diverse hardware and application constraints specific to edge platforms remains a significant challenge. In this paper, we introduce a zero-shot NAS approach designed to generate hardware-aware architectures, combined with a selection technique that allows adaptable model optimization across various deployment scenarios without re-executing the search. We demonstrate the flexibility of our solution by benchmarking the Google Coral Edge Tensor Processing Unit (TPU). Our technique led to the efficient exploration of the architecture space of NAS-Bench-201 (NB201) in under a minute, accelerating the search by 25x compared to previous work while maintaining comparable accuracies of 93.24% on CIFAR-10 and 42.99% on ImageNet16-120.
Unearthing nanoplastics in soil: optimising extraction and purification while preserving particle integrity
Item type: Journal Article
Forsyth, Hannah; Gnoffo, Chiara; Oh, Sehui; et al. (2026)
Nanoplastics (NPs) are considered to be widespread environmental pollutants but little is known about their occurrence and properties in soils. Here, we evaluate and optimise an extraction and purification method for NPs in soil, aiming to preserve particle integrity and assess the potential for a single extraction workflow to support characterisation by both advanced microscopy techniques and mass-based techniques such as Py-GC-MS. This targets comprehensive characterisation of NPs, including size, shape, polymer chemistry, and mass concentration data. Individual extraction and purification steps were optimised, including density separation by centrifugation using a sucrose solution, filtration to < 1 μm by vacuum filtration and concentration combined with purification using direct flow ultrafiltration. Recovery tests using Pd-doped NPs aided baseline performance quantification. Recoveries were 38% for extraction (7% SD), 74% for density separation (18% SD), 92% for filtration (15% SD), and 74% for ultrafiltration (7% SD). The final combined method comprising these steps in sequence had a low recovery of 1.4% (0.4% SD), demonstrating the challenge of particle-preserving NPs extraction from soil. Next, we assessed the suitability of our final combined NP extraction method for analysis of NPs by Py-GC-MS and SEM. As part of this feasibility assessment, we tested a step to transfer the extracted NPs for Py-GC-MS analysis by dissolving them in a mixture of 1,2,4-trichlorobenzene and p-xylene and drying aliquots in pyrolysis cups (87% recovery, 41% SD). However, high sample dilution during the extraction resulted in a high method detection limit, which was unsuitable for quantitative Py-GC-MS analysis of NPs extracted from soil. In contrast, the method was well suited to qualitative analysis of NPs using advanced microscopy techniques, with SEM images revealing highly purified samples and minimal contamination from soil. As the first study to both evaluate recovery for a particle-preserving NP extraction method in soil and demonstrate its application to analytical techniques, this work provides a foundation for future improvements in NP extraction and analysis methods.