Tue use of BIM-modeling (BIM) within the building life cycle depends on different (specialist) models as well as a detailed as-built-model. The legally compliant transfer of data and models is decisive for the vision of a consistent use of BIM. Past legal research has mainly dealt with the question of copy and IP rights in contracts. The new EU legal framework of data law is now a game changer as codified legal rules for data use are on their way. The paper will reflect these requirements and link them to the relevant ISO norms. The paper will show how legal governance of BIM-models can be designed by a consistent model rights management which is linked to the technical delivery of information and present guidelines for a legally compliant transfer of BIM-models and data, promoting the trustworthy sharing of data and models within the building life cycle.

In this work we are describing our preliminary experimental evaluation of open-source machine learning based pose estimation solutions for tracking a surfer’s skeleton on a stationary wave. We are briefly describing the sport’s environment and development and the current state-of-the-art of machine learning based single camera 2D and 3D pose estimation solutions. The main part of the paper deals with the experiment’s setup and the interpretation of the resulting data of the movements of one surfer. We are closing with our lessons learned and our next steps. 

Das Thema Künstliche Intelligenz (KI) findet auch im Kulturmarketing immer mehr Anwendung und bietet dabei vielfältige Optimierungsmöglichkeiten. Welche Chancen und Herausforderungen damit für die operativen Aufgaben, Prozesse und notwendigen Kompetenzen im Kulturmarketing einhergehen, erklärt uns Andreas Wagener, Professor für E-Commerce und Social Media an der Hochschule Hof.

Im Zuge der Covid-Pandemie hat sich insbesondere in China der Trend des “live-eCommerce“ verstärkt, der dort im Jahr 2022 auf ein Volumen von 43,2 Milliarden Yuan geschätzt wird (etwa €6Mrd.) Dieser lässt sich in etwa als „digitales Teleshopping“ umschreiben und umfasst oft eine intensive Interaktion mit dem Publikum. Zuschauer können „live“ Fragen zu präsentierten Produkten stellen. Grundsätzlich lässt sich dies auch in den virtuellen Welten realisieren, wie zum Beispiel der chinesische Handelsriese AliBaba mit seinem Ansatz auf der Plattform TaoBao zeigt. Dort entstehen eigene dreidimensionale virtuelle Shopping-Umgebungen, die per Avatar besuchbar und „erlebbar“ sind. Die einzelnen Verkaufsabteilungen und Geschäfte sind mit – virtualisierten – Mitarbeitern besetzt, welche die Kunden in „Echtzeit“ beraten. Zu festen Terminen sollen zudem – analog zu dem bereits praktizierten „gewöhnlichen“ online-Live-Shopping – interaktive Verkaufsveranstaltungen und Produktpräsentation stattfinden.

Contact Tracing Apps (CTAs) have been developed to contain the coronavirus disease 19 ( COVID-19) spread. By design, such apps invade their users’ privacy by recording data about their health, contacts, and—partially—location. Many CTAs frequently broadcast pseudorandom numbers via Bluetooth to detect encounters. These numbers are changed regularly to prevent individual smartphones from being trivially trackable. However, the effectiveness of this procedure has been little studied.

We measured real smartphones and observed that the German Corona-Warn-App (CWA ) exhibits a device-specific latency between two subsequent broadcasts. These timing differences provide a potential attack vector for fingerprinting smartphones by passively recording Bluetooth messages. This could conceivably lead to the tracking of users’ trajectories and, ultimately, the re-identification of users.

The memory controller of the CPU uses bank addressing functions to determine physical locations within DRAM DIMMs. There are many fields of application for these addressing functions, particularly in security. For exam- ple, many Rowhammer proofs-of-concept use bank addressing functions to select addresses located on the same bank but in different rows to produce row conflicts. AMD provides these addressing functions for older CPU models. Hence, research on reverse-engineering addressing functions mainly targeted Intel CPUs since Intel did not publish these functions. However, AMD stopped to publish the DRAM addressing functions several years ago. AMD manufactures roughly a third of the sold CPUs in today’s CPU market. We analyze reverse- engineering tools for addressing functions and find that they do not work with AMD CPUs, hindering reverse-engineering at- tempts and Rowhammer attacks on systems with AMD CPUs. In this paper, we introduce an approach to reverse-engineer the addressing functions of AMD CPUs, which facilitates future Rowhammer experiments on AMD CPUs.

Large language models (LLMs) have garnered significant attention, but the definition of "large" lacks clarity. This paper focuses on medium-sized lan-guage models (MLMs), defined as having at least six billion parameters but less than 100 billion. The study evaluates MLMs regarding zero-shot genera-tive question answering, which requires models to provide elaborate answers without external document retrieval. The paper introduces an own test da-taset and presents results from human evaluation. Results show that combin-ing the best answers from different MLMs yielded an overall correct answer rate of 82.7% which is better than the 60.9% of ChatGPT. The best MLM achieved 46.4% and has 7B parameters, which highlights the importance of using appropriate training data for fine-tuning rather than solely relying on the number of parameters. More fine-grained feedback should be used to further improve the quality of answers.

Background

Most migraine patients need an effective acute medication. Real-world data can provide important information on the performance of acute migraine medication in clinical practice.
Methods
We used data from the German Migraine and Headache Society Headache Registry, where patients rate efficacy and tolerability of and satisfaction with each of their acute headache medications.
Results
A total of 1756 adult migraine patients (females: 85%, age: 39.5 ± 12.8 years, headache days per month: 13.5 ± 8.1) were included. Of these, 93% used acute medication, most frequently triptans (59.3%) and/or non-opioid analgesics (56.4%), and 58.5% rated efficacy as good or very good. This was more frequent for triptans (75.4%) than for non-opioid analgesics (43.6%, p < 0.001). Among non-opioid analgesics, naproxen was rated most effective (61.9% very good or good, p < 0.001 compared to ibuprofen, acetylsalicylic acid and paracetamol). Patient-rated efficacy significantly declined with higher headache frequencies (p < 0.001), and this effect remained significant after omitting patients overusing acute medication.
Conclusion
In the present population recruited at specialized headache centers, patients rated triptans as more effective than non-opioid analgesics, naproxen as more effective than ibuprofen, and acute medication efficacy decreased with increasing headache frequency.

Trial registration: The German Migraine and Headache Society Headache Registry is registered with the German Clinical Trials Register (DRKS 00021081).

Der Hype um KI-Werkzeuge wie ChatGPT, Dall-E oder Midjourney ist ungebrochen. Ohne Frage wird generative KI das Marketing ganz grundsätzlich verändern. Unser Experte Andreas Wagener zeigt, wo KI bereits Anwendung findet und wie Sie sie nutzen können.