Modern DRAM is susceptible to fault attacks that undermine the entire system’s security. The most well-studied disturbance effect is Rowhammer, where an attacker repeatedly opens and closes (i.e., hammers) different rows, which can lead to bitflips in adjacent rows. Different hammering strategies include double-sided, hammering two rows sandwiching a victim row, and one-location, hammering a single row. One-location Rowhammer requires no physical address information, as any location in memory is mapped to a DRAM row, and no relation between rows is required for hammering. The recently discovered Rowpress differs from Rowhammer by not hammering rows but keeping them open longer, evident by a disjoint set of affected memory locations.

In this paper, we examine the differences between four attack variants: one-location Rowhammer, a one-location Rowpress variant we developed, double-sided Rowhammer, and double-sided Rowpress on a set of 12 DDR4 modules. Our methodology is to hammer and press the exact same set of physical memory locations in all attack variants. Surprisingly, our results show that on 4 out of 12 DDR4 modules, we were only able to reproduce double-sided Rowhammer but none of the other attack variants. On 2 DDR4 modules, we were able to reproduce all attack variants. We find that the number of unique bitflip locations ranges from 161 to 15 612, when hammering the exact same set of physical memory locations. Our one-location Rowhammer attack induces roughly the same amount of bitflips as double-sided Rowhammer, however, only 61.8 % of bitflip locations overlap. We explain this by one-location Rowhammer inducing bitflips due to the Rowhammer as well as the Rowpress effect, making the differentiation of both methods difficult, therefore, calling it Presshammer. Based on our observed bitflips, we develop the first end-to-end one-location Rowpress attack. One-location Rowpress requires only minimal physical address information that an attacker can acquire through a same-row same-bank side-channel attack. Our end-to- end attack escalates to kernel privileges within less than 10 minutes.

Large language models (LLMs) have garnered significant attention, but the definition of “large” lacks clarity. This paper focuses on medium-sized language models (MLMs), defined as having at least six billion parameters but less than 100 billion. The study evaluates MLMs regarding zero-shot generative question answering in German and English language, which requires models to provide elaborate answers without external document retrieval (RAG). The paper introduces an own test dataset and presents results from human evaluation. Results show that combining 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 English MLM achieved 71.8% and has 33B parameters, which highlights the importance of using appropriate training data for fine-tuning rather than solely relying on the number of parameters. The best German model also surpasses ChatGPT for the equivalent dataset. More fine-grained feedback should be used to further improve the quality of answers. The open source community is quickly closing the gap to the best commercial models.

This book summarizes studies conducted by the Empirical Research and User Experience (ERUX) research group at the Institute for Information Systems (iisys) at Hof University of Applied Sciences over the period 2019 to 2022. The main topic focuses on “The Human-Machine-Interface in Advanced Driver Assistance Systems towards Autonomous Driving”, investigated using the example of automatic parking assistant in cars. A broad spectrum of methods was used to ensure comprehensive analyses and valid results. Starting situation for the studies has been the industry's realization that the demand for parking assistants had fallen short of expectations. Moreover, they concluded that a host of driving assistant system owners´ do not use them. This indicates an unsatisfactory acceptance of new technologies on the part of car drivers. The aim of the studies was therefore to determine the reasons for this restraint, as well as to derive measures to increase acceptance of this systems among end customers.

The first study focused on operators and the restraint of German drivers towards driver assistance systems. The results of this quantitative study displayed that although the modernity of these systems has been identified, the functional enhancements they offer are insufficient. In order to overcome this discrepancy, manufacturers are advised to emphasize the functional advantages more strongly when communication with the end customer. Study also suggested that direct experience can significantly increase customers' willingness to purchase such a system. Therefore, it is of pivotal importance to introduce customers to the new technologies in a targeted manner.

Using conjoint measurement, a second study illustrates driver assistance systems already representing a significant benefit when purchasing a car, even though the trust in fully autonomous vehicles is still restricted. Drivers favor established car manufacturer in terms of the development of autonomous vehicles. The third study utilizes eye-tracking to examine operating patterns of parking assistance systems. Moreover, it derives potential for improvement in terms of intuitive design and user experience. Fourthly, alternative scenarios for the design of parking assistants are evaluated in a concept test. This indicated the vital importance of simple and acquainted operating options.

Across all methods, the customer actually being able to experience driving assistance systems is a decisive matter for the acceptance of these new technologies. Therefore, manufacturers should emphasize their focus on getting customers in touch with these systems, as well as offering them targeted usage options. In addition, the integration of customer feedback into development efforts is essential to produce intuitive, user-friendly user interfaces and thus promote the diffusion process of new technologies.