德国联邦物理技术研究院

Low-Gain Avalanche Diodes (LGADs) are fast silicon sensors with internal charge multiplication and are key candidates for precision timing layers in future high-energy hadron colliders. Their operation in harsh radiation environments, however, is limited by acceptor removal in the gain layer, which reduces the active acceptor concentration and degrades the internal electric field required for avalanche multiplication. Improving the radiation tolerance of the gain layer is therefore essential for future 4D tracking applications. In this work, we investigated several LGAD prototypes produced in collaboration with Hamamatsu Photonics K.K. (HPK), featuring modified gain-layer designs, including oxygen-modified, carbon-implanted, and boron–phosphorus compensated structures. The sensors were studied after proton and reactor-neutron irradiation. Radiation tolerance was characterized using the acceptor-removal coefficient extracted from IV measurements and the operation voltage required to recover the timing performance after irradiation. The results show that carbon implantation is the only approach among those studied here that provides a clear improvement in radiation tolerance. In contrast, neither oxygen-related modification, including the Partially Activated Boron (PAB) approach, nor gain-layer compensation alone yields a significant improvement, and the compensated carbon-implanted structure shows no clear advantage over the carbon-only case. In addition, the acceptor-removal coefficient is found to depend on the irradiation particle type and energy.

Developments in ultrastable lasers have fueled remarkable advances in optical frequency metrology and quantum science. A key ingredient in further improving laser frequency stability is the use of low-noise mirror materials such as AlGaAs crystalline coatings. However, excess noise observed with these coatings limits the performance of cryogenic silicon cavities with AlGaAs mirrors to similar levels achieved with conventional dielectric coatings. With a new pair of crystalline coated mirrors in a 6-cm-long cryogenic silicon cavity operated at 17 K, we demonstrate a clear advantage of crystalline coatings over dielectric coatings. The achieved fractional frequency stability of 2.5×10^{-17} at 10 s is four times better than expected for dielectric mirrors and corresponds to more than a tenfold reduction in the coating mechanical loss factor. We also combine two silicon cavities to demonstrate optical frequency averaging for enhanced stability. In addition, we present a long-term frequency drift record of four cryogenic silicon cavities measured over several years. These results open up realistic prospects for cavity-stabilized lasers with 10^{-18} fractional stability, as well as an all-optical timescale with continuously operating optical local oscillators.

Photoplethysmography (PPG) is a widely used non-invasive physiological sensing technique, suitable for various clinical applications. Such clinical applications are increasingly supported by machine learning methods, raising the question of the most appropriate input representation and model choice. Comprehensive comparisons, in particular across different input representations, are scarce. We address this gap in the research landscape by a comprehensive benchmarking study covering three kinds of input representations, interpretable features, image representations and raw waveforms, across prototypical regression and classification use cases: blood pressure and atrial fibrillation prediction. In both cases, the best results are achieved by deep neural networks operating on raw time series as input representations. Within this model class, best results are achieved by modern convolutional neural networks (CNNs). but depending on the task setup, shallow CNNs are often also very competitive. We envision that these results will be insightful for researchers to guide their choice on machine learning tasks for PPG data, even beyond the use cases presented in this work.

Large pre-trained language models have become a crucial backbone for many downstream tasks in natural language processing (NLP), and while they are trained on a plethora of data containing a variety of biases, such as gender biases, it has been shown that they can also inherit such biases in their weights, potentially affecting their prediction behavior. However, it is unclear to what extent these biases also affect feature attributions generated by applying "explainable artificial intelligence" (XAI) techniques, possibly in unfavorable ways. To systematically study this question, we create a gender-controlled text dataset, GECO, in which the alteration of grammatical gender forms induces class-specific words and provides ground truth feature attributions for gender classification tasks. This enables an objective evaluation of the correctness of XAI methods. We apply this dataset to the pre-trained BERT model, which we fine-tune to different degrees, to quantitatively measure how pre-training induces undesirable bias in feature attributions and to what extent fine-tuning can mitigate such explanation bias. To this extent, we provide GECOBench, a rigorous quantitative evaluation framework for benchmarking popular XAI methods. We show a clear dependency between explanation performance and the number of fine-tuned layers, where XAI methods are observed to benefit particularly from fine-tuning or complete retraining of embedding layers.

Entangled states hold the promise of improving the precision and accuracy of quantum sensors. We experimentally demonstrate that spectroscopy of an optical clock transition using entangled states can outperform its classical counterpart. Two ^{40}Ca^{+} ions are entangled in a quantum state with vanishing first-order magnetic field sensitivity, extending the coherence time of the atoms and enabling near-lifetime-limited probe times of up to 550 ms. In our protocol, entangled ions reach the same instability as uncorrelated ions, but at half the probe time, enabling faster cycle times of the clock. We run two entangled ^{40}Ca^{+} ions as an optical clock and compare its frequency instability with a ^{87}Sr lattice clock. The instability of the entangled ion clock is below a clock operated with classically correlated states for all probe times. We observe instabilities below the theoretically expected quantum projection noise limit of two uncorrelated ions for interrogation times below 100 ms. The lowest fractional frequency instability of 7×10^{-16}/sqrt[τ/1  s] is reached for 250 ms probe time, limited by residual phase noise of the probe laser. This represents the lowest instability reported to date for a ^{40}Ca^{+} ion clock.