The National Library of Luxembourg (French: Bibliothèque nationale du Luxembourg), abbreviated as BnL, is Luxembourg's national library. It was founded in its current form in 1899, as a result of a series of different institutions originating in the 18th century. It is located in the Kirchberg district of Luxembourg City. The BnL is a public establishment under the supervision of the Ministry of Culture.The BnL holds 1.8 million printed items, making it the largest library in Luxembourg. The library’s collections include both print and digital documents, such as books, manuscripts, journals, newspapers, magazines, databases, maps, stamps, prints, drawings and scores by Luxembourgish composers. Roughly three quarters of its contents, particularly scientific resources, come from abroad.As a legal deposit library, the BnL receives copies of books and other printed and digital documents published in Luxembourg. It is also the country’s national ISBN, ISSN, ISMN and ISNI agency.
The trigger and data acquisition systems of the LHC experiments rely on reconfigurable architectures based on FPGA (Field Programmable Gate Arrays) to accommodate design updates and increasing of the detectors complexity. However, this flexibility comes at the cost of longer development and validation cycles, particularly for large real-time systems operating under strict latency and throughput constraints. This work presents the HLS-based firmware redesign of the back-end system from the ATLAS Liquid Argon Calorimeter, where the introduction of a parallel switch matrix on REMAP block significantly simplifies the architecture and reduces the use of firmware resources. The proposed methodology allows systematic design exploration without modifying the source code, shortens verification time, and provides a more efficient workflow for creating and maintaining large-scale FPGA firmware in modern trigger systems.
Caribou is a versatile data acquisition (DAQ) system developed within several collaborative frameworks (CERN EP R D, DRD3, AIDAinnova, and Tangerine) to support laboratory and test-beam characterization of novel silicon pixel detectors. It combines a custom Control and Readout (CaR) board with a Xilinx Zynq System-on-Chip (SoC) running project-wide shared firmware and software stacks. The system architecture emphasizes reusability, flexibility, and ease of integration. The CaR board provides essential interfaces such as programmable power supplies, voltage and current references, high-speed ADCs, and configurable I/O lines for detector control and readout. The SoC runs an embedded Linux distribution built with PetaLinux and integrates two main components: Peary, a C++ embedded DAQ application providing hardware abstraction, configuration management, logging, and multi-device control through Command Line (CLI) and Python interfaces; and Boreal, a common Caribou FPGA firmware framework offering reusable modules and automated build workflows for user-specific bit files. The next major milestone in Caribou's evolution is the transition to version 2.0, based on a Zynq UltraScale+ System-on-Module (SoM) architecture. This paper presents the recent progress and future prospects of the project and describes recent hardware, firmware, and software developments preparing the system for the upcoming CaR board v2.0.
We measure the time-integrated CP asymmetry, A _CP , in D^0 → K^0_S K^0_S decays reconstructed in e^+e^-→ cc̅ events collected by the Belle and Belle II experiments with the data sample corresponding to an integrated luminosities of 980 fb^-1 and 427 fb^-1 , respectively. The D^0 decays are required to come from the flavor-conserving D^*+→ D^0π ^+ decay ( D^*+ tag) to determine the charm flavor at production time. A control sample of D^*+ -tagged D^0→ K^+ K^- is used to correct for production and detection asymmetries. The result, A_CP(D^0→ K^0_S K^0_S ) = (-1.4 ± 1.3(stat) ± 0.1(syst))% , is consistent with previous determinations and with CP symmetry.
Search query understanding (QU) is an important building block of the modern e-commerce search engines. QU extracts multiple intents from customer queries, including intended color, brand, etc. One of the most important tasks in QU is predicting which product category the user is interested in. In our work we are tapping into query product type classification (Q2PT) task. Compared to classification of full-fledged texts, Q2PT is more complicated because of the ambiguity of short search queries, which is aggravated by language and cultural differences in worldwide online stores. Moreover, the span and variety of product categories in modern marketplaces pose a significant challenge. We focus on Q2PT inference in the global multi-locale e-commerce markets, which need to deliver high quality user experience in both large and small local stores alike. The common approach of training Q2PT models for each locale separately shows significant performance drops in low-resource stores and prevents from easily expanding to a new country, where the Q2PT model has to be created from scratch. We use transfer learning to address this challenge, augmenting low-resource locales through the vast knowledge of the high-resource ones. We introduce a unified, locale-aware Q2PT model, sharing training data and model structure across worldwide stores. We show that the proposed unified locale-aware Q2PT model has superior performance over the alternatives by conducting extensive quantitative and qualitative analysis on the large-scale multilingual e-commerce dataset across 20 worldwide locales. Our online A/B tests have shown that using locale-aware model improves over the previous user experience, increasing customer satisfaction.
