Search

Internal features from large-scale pre-trained diffusion models have recently been established as powerful semantic descriptors for a wide range of downstream tasks. Works that use these features generally need to add noise to images before passing them through the model to obtain the semantic features, as the models do not offer the most useful features when given images with little to no noise. We show that this noise has a critical impact on the usefulness of these features that cannot be remedied by ensembling with different random noises. We address this issue by introducing a lightweight, unsupervised fine-tuning method that enables diffusion backbones to provide high-quality, noise-free semantic features. We show that these features readily outperform previous diffusion features by a wide margin in a wide variety of extraction setups and downstream tasks, offering better performance than even ensemble-based methods at a fraction of the cost., Comment: for the project page and code, view https://compvis.github.io/CleanDIFT/

As supercomputers grow in hardware complexity, their susceptibility to faults increases and measures need to be taken to ensure the correctness of results. Some numerical algorithms have certain characteristics that allow them to recover from some types of faults. It has been demonstrated that adaptive Runge-Kutta methods provide resilience against transient faults without adding computational cost. Using recent advances in adaptive step size selection for spectral deferred correction (SDC), an iterative numerical time stepping scheme that can produce methods of arbitrary order, we show that adaptive SDC can also detect and correct transient faults. Its performance is found to be comparable to that of the dedicated resilience strategy Hot Rod.

Accurate tire modeling is crucial for optimizing autonomous racing vehicles, as state-of-the-art (SotA) model-based techniques rely on precise knowledge of the vehicle's parameters. Yet, system identification in dynamic racing conditions is challenging due to varying track and tire conditions. Traditional methods require extensive operational ranges, often impractical in racing scenarios. Machine learning (ML)-based methods, while improving performance, struggle with generalization and depend on accurate initialization. This paper introduces a novel on-track system identification algorithm, incorporating a neural network (NN) for error correction, which is then employed for traditional system identification with virtually generated data. Crucially, the process is iteratively reapplied, with tire parameters updated at each cycle, leading to notable improvements in accuracy in tests on a scaled vehicle. Experiments show that it is possible to learn a tire model without prior knowledge with only 30 seconds of driving data and 3 seconds of training time. This method demonstrates greater one-step prediction accuracy than the baseline nonlinear least squares (NLS) method under noisy conditions, achieving a 3.3x lower root mean square error (RMSE), and yields tire models with comparable accuracy to traditional steady-state system identification. Furthermore, unlike steady-state methods requiring large spaces and specific experimental setups, the proposed approach identifies tire parameters directly on a race track in dynamic racing environments.

How can robots learn and adapt to new tasks and situations with little data? Systematic exploration and simulation are crucial tools for efficient robot learning. We present a novel black-box policy search algorithm focused on data-efficient policy improvements. The algorithm learns directly on the robot and treats simulation as an additional information source to speed up the learning process. At the core of the algorithm, a probabilistic model learns the dependence of the policy parameters and the robot learning objective not only by performing experiments on the robot, but also by leveraging data from a simulator. This substantially reduces interaction time with the robot. Using this model, we can guarantee improvements with high probability for each policy update, thereby facilitating fast, goal-oriented learning. We evaluate our algorithm on simulated fine-tuning tasks and demonstrate the data-efficiency of the proposed dual-information source optimization algorithm. In a real robot learning experiment, we show fast and successful task learning on a robot manipulator with the aid of an imperfect simulator.

The numerical method of dynamical low-rank approximation (DLRA) has recently been applied to various kinetic equations showing a significant reduction of the computational effort. In this paper, we apply this concept to the linear Boltzmann-Bhatnagar-Gross-Krook (Boltzmann-BGK) equation which due its high dimensionality is challenging to solve. Inspired by the special structure of the non-linear Boltzmann-BGK problem, we consider a multiplicative splitting of the distribution function. We propose a rank-adaptive DLRA scheme making use of the basis update & Galerkin integrator and combine it with an additional basis augmentation to ensure numerical stability, for which an analytical proof is given and a classical hyperbolic Courant-Friedrichs-Lewy (CFL) condition is derived. This allows for a further acceleration of computational times and a better understanding of the underlying problem in finding a suitable discretization of the system. Numerical results of a series of different test examples confirm the accuracy and efficiency of the proposed method compared to the numerical solution of the full system.

Laser-structuring techniques like Direct Laser Interference Patterning show great potential for optimizing electrodes for water electrolysis. Therefore, a systematic experimental study based on statistical design of experiments is performed to analyze the influence of the spatial period and the aspect ratio between spatial period and structure depth on the electrode performance for pure Ni electrodes. The electrochemically active surface area could be increased by a factor of 12 compared to a non-structured electrode. For oxygen evolution reaction, a significantly lower onset potential and overpotential ($\approx$-164 mV at 100 mA/cm$^2$) is found. This is explained by a lower number of active nucleation sites and, simultaneously, larger detached bubbles, resulting in reduced electrode blocking and thus, lower ohmic resistance. It is found that the spatial distance between the laser-structures is the decisive processing parameter for the improvement of the electrode performance., Comment: Manuscript: 27 pages, 10 figures, 4 tables; Electronic Supplementary Information: 13 pages, 11 figures, 3 tables

Recently, an interesting pattern was found in the differential equations satisfied by the Feynman integrals describing tree-level correlators of conformally coupled scalars in a power-law FRW cosmology [1,2]. It was proven that simple and universal graphical rules predict the equations for arbitrary graphs as a flow in kinematic space. In this note, we show that the same rules$\unicode{x2013}$with one small addition$\unicode{x2013}$also determine the differential equations for loop integrands. We explain that both the basis of master integrals and the singularities of the differential equations can be represented by tubings of marked graphs. An important novelty in the case of loops is that some basis functions can vanish, and we present a graphical rule to identify these vanishing functions. Taking this into account, we then demonstrate that the kinematic flow correctly predicts the differential equations for all loop integrands., Comment: 32 pages

Advances in lightweight neural networks have revolutionized computer vision in a broad range of IoT applications, encompassing remote monitoring and process automation. However, the detection of small objects, which is crucial for many of these applications, remains an underexplored area in current computer vision research, particularly for low-power embedded devices that host resource-constrained processors. To address said gap, this paper proposes an adaptive tiling method for lightweight and energy-efficient object detection networks, including YOLO-based models and the popular FOMO network. The proposed tiling enables object detection on low-power MCUs with no compromise on accuracy compared to large-scale detection models. The benefit of the proposed method is demonstrated by applying it to FOMO and TinyissimoYOLO networks on a novel RISC-V-based MCU with built-in ML accelerators. Extensive experimental results show that the proposed tiling method boosts the F1-score by up to 225% for both FOMO and TinyissimoYOLO networks while reducing the average object count error by up to 76% with FOMO and up to 89% for TinyissimoYOLO. Furthermore, the findings of this work indicate that using a soft F1 loss over the popular binary cross-entropy loss can serve as an implicit non-maximum suppression for the FOMO network. To evaluate the real-world performance, the networks are deployed on the RISC-V based GAP9 microcontroller from GreenWaves Technologies, showcasing the proposed method's ability to strike a balance between detection performance ($58% - 95%$ F1 score), low latency (0.6 ms/Inference - 16.2 ms/Inference}), and energy efficiency (31 uJ/Inference} - 1.27 mJ/Inference) while performing multiple predictions using high-resolution images on a MCU., Comment: arXiv admin note: text overlap with arXiv:2311.07163

There exists a dictionary between hereditary orders and smooth stacky curves, resp. tame orders of global dimension 2 and Azumaya algebras on smooth stacky surfaces. We extend this dictionary by explaining how the restriction of a tame order to a curve on the underlying surface corresponds to the fiber product of the curve with the stacky surface. By considering "bad" intersections we can start extending the dictionary in the 1-dimensional case to include non-hereditary orders and singular stacky curves. Two applications of these results are a novel description and classification of noncommutative conics in graded Clifford algebras, giving a geometric proof of results of Hu-Matsuno-Mori, and a complete understanding and classification of skew cubics, generalizing the work of Kanazawa for Fermat skew cubics., Comment: 41 pages, all comments welcome

Head-to-head racing against opponents is a challenging and emerging topic in the domain of autonomous racing. We propose Predictive Spliner, a data-driven overtaking planner that learns the behavior of opponents through Gaussian Process (GP) regression, which is then leveraged to compute viable overtaking maneuvers in future sections of the racing track. Experimentally validated on a 1:10 scale autonomous racing platform using Light Detection and Ranging (LiDAR) information to perceive the opponent, Predictive Spliner outperforms State-of-the-Art (SotA) algorithms by overtaking opponents at up to 83.1% of its own speed, being on average 8.4% faster than the previous best-performing method. Additionally, it achieves an average success rate of 84.5%, which is 47.6% higher than the previous best-performing method. The method maintains computational efficiency with a Central Processing Unit (CPU) load of 22.79% and a computation time of 8.4 ms, evaluated on a Commercial off-the-Shelf (CotS) Intel i7-1165G7, making it suitable for real-time robotic applications. These results highlight the potential of Predictive Spliner to enhance the performance and safety of autonomous racing vehicles. The code for Predictive Spliner is available at: https://github.com/ForzaETH/predictive-spliner., Comment: Accepted to RA-L

We propose cognitive prompting as a novel approach to guide problem-solving in large language models (LLMs) through structured, human-like cognitive operations, such as goal clarification, decomposition, filtering, abstraction, and pattern recognition. By employing systematic, step-by-step reasoning, cognitive prompting enables LLMs to tackle complex, multi-step tasks more efficiently. We introduce three variants: a deterministic sequence of cognitive operations, a self-adaptive variant in which the LLM dynamically selects the sequence of cognitive operations, and a hybrid variant that uses generated correct solutions as few-shot chain-of-thought prompts. Experiments with LLaMA, Gemma~2, and Qwen models in each two sizes on the arithmetic reasoning benchmark GSM8K demonstrate that cognitive prompting significantly improves performance compared to standard question answering., Comment: 6 pages, submitted to ESANN 2025

The remarkable ecological success of humans is often attributed to our ability to develop complex cultural artefacts that enable us to cope with environmental challenges. The evolution of complex culture (cumulative cultural evolution) is usually modeled as a collective process in which individuals invent new artefacts (innovation) and copy information from others (social learning). This classic picture overlooks the fact that in the digital age, intelligent algorithms are increasingly mediating information between humans, with potential consequences for cumulative cultural evolution. Building on an established model of cultural evolution, we investigate the combined effects of network-based social learning and a simplistic version of algorithmic mediation on cultural accumulation. We find that algorithmic mediation has a strong impact on cultural accumulation and that this impact generally increases as social networks become less densely connected. Moreover, cultural accumulation tends to be optimal when social learning and algorithmic mediation are combined, and the optimal ratio depends on the network's density. Our modeling results are a first step towards formalising the impact of intelligent algorithms on cumulative cultural evolution within an established framework. Models of this kind will also help to uncover mechanisms of human-machine interaction in cultural contexts, guiding hypotheses for future experimental testing.

In this article, we introduce an original hybrid quantum-classical algorithm based on a variational quantum algorithm for solving systems of differential equations. The algorithm relies on a spectral method, which involves encoding the solution functions in the amplitudes of the quantum states generated by different parametrized circuits and transforms the task of solving the differential equations into an optimization problem. We first describe the principle of the algorithm from a theoretical point of view. We provide a detailed pseudo-code of the algorithm, on which we conduct a complexity analysis to highlight its scaling properties. We apply it to a set of examples, showcasing its applicability across diverse sets of differential equations. We discuss the advantages of our method and potential avenues for further exploration and refinement., Comment: The algorithm described in this paper is subject to a patent submission (EP24306601). Use of this method requires a license, which can be obtained from ColibriTD

Basket trials examine the efficacy of an intervention in multiple patient subgroups simultaneously. The division into subgroups, called baskets, is based on matching medical characteristics, which may result in small sample sizes within baskets that are also likely to differ. Sparse data complicate statistical inference. Several Bayesian methods have been proposed in the literature that allow information sharing between baskets to increase statistical power. In this work, we provide a systematic comparison of five different Bayesian basket trial designs when sample sizes differ between baskets. We consider the power prior approach with both known and new weighting methods, a design by Fujikawa et al., as well as models based on Bayesian hierarchical modeling and Bayesian model averaging. The results of our simulation study show a high sensitivity to changing sample sizes for Fujikawa's design and the power prior approach. Limiting the amount of shared information was found to be decisive for the robustness to varying basket sizes. In combination with the power prior approach, this resulted in the best performance and the most reliable detection of an effect of the treatment under investigation and its absence.

Hyperspectral imaging (HSI) is emerging as a promising novel imaging modality with various potential surgical applications. Currently available cameras, however, suffer from poor integration into the clinical workflow because they require the lights to be switched off, or the camera to be manually recalibrated as soon as lighting conditions change. Given this critical bottleneck, the contribution of this paper is threefold: (1) We demonstrate that dynamically changing lighting conditions in the operating room dramatically affect the performance of HSI applications, namely physiological parameter estimation, and surgical scene segmentation. (2) We propose a novel learning-based approach to automatically recalibrating hyperspectral images during surgery and show that it is sufficiently accurate to replace the tedious process of white reference-based recalibration. (3) Based on a total of 742 HSI cubes from a phantom, porcine models, and rats we show that our recalibration method not only outperforms previously proposed methods, but also generalizes across species, lighting conditions, and image processing tasks. Due to its simple workflow integration as well as high accuracy, speed, and generalization capabilities, our method could evolve as a central component in clinical surgical HSI., Comment: Oral at MICCAI 2024

We investigated the electronic properties of the topological insulator Bi$_{2}$Te$_{3}$ by scanning tunneling microscopy and spectroscopy at low temperature. We obtained high-resolution quasiparticle interference data of the topological surface Dirac electrons at different energies. Spin-selective joint density of states calculations were performed for surface and bulk electronic states to interpret the observed quasiparticle interference data. The topological properties of our crystals are demonstrated by the absence of backscattering along with the linear energy dispersion of the dominant scattering vector. In addition, we detect non-dispersive scattering modes which we associate with bulk-surface scattering and, thus, allow an approximate identification of the bulk energy gap range based on our quasiparticle interference data. Measurements of differential conductance maps in magnetic fields up to 15 T have been carried out, but no strong modifications could be observed., Comment: 14 pages, 22 figures

Evolutionary algorithms (EAs) are universal solvers inspired by principles of natural evolution. In many applications, EAs produce astonishingly good solutions. As they are able to deal with complex optimisation problems, they show great promise for hard problems encountered in the field of graph drawing.To complement recent theoretical advances in the analysis of EAs on graph drawing, we contribute a fundamental empirical study. We consider the so-called \textsc{One-Sided Bipartite Crossing Minimisation (OBCM)}: given two layers of a bipartite graph and a fixed horizontal order of vertices on the first layer, the task is to order the vertices on the second layer to minimise the number of edge crossings. We empirically analyse the performance of simple EAs for OBCM and compare different mutation operators on the underlying permutation ordering problem: exchanging two elements (\textit{exchange}), swapping adjacent elements (\textit{swap}) and jumping an element to a new position (\textit{jump}). EAs using jumps easily outperform all deterministic algorithms in terms of solution quality after a reasonable number of generations. We also design variations of the best-performing EAs to reduce the execution time for each generation. The improved EAs can obtain the same solution quality as before and run up to 100 times faster., Comment: 4 pages, 3 figures, 1 poster. To appear in the 32nd International Symposium on Graph Drawing and Network Visualization

Safe Bayesian optimization (BO) algorithms promise to find optimal control policies without knowing the system dynamics while at the same time guaranteeing safety with high probability. In exchange for those guarantees, popular algorithms require a smoothness assumption: a known upper bound on a norm in a reproducing kernel Hilbert space (RKHS). The RKHS is a potentially infinite-dimensional space, and it is unclear how to, in practice, obtain an upper bound of an unknown function in its corresponding RKHS. In response, we propose an algorithm that estimates an upper bound on the RKHS norm of an unknown function from data and investigate its theoretical properties. Moreover, akin to Lipschitz-based methods, we treat the RKHS norm as a local rather than a global object, and thus reduce conservatism. Integrating the RKHS norm estimation and the local interpretation of the RKHS norm into a safe BO algorithm yields PACSBO, an algorithm for probably approximately correct safe Bayesian optimization, for which we provide numerical and hardware experiments that demonstrate its applicability and benefits over popular safe BO algorithms., Comment: Accepted to the Symposium on Systems Theory in Data and Optimization (SysDO 2024). This is a preprint of the final version, which is to appear in Lecture Notes in Control and Information Sciences - Proceedings

The Quadratic Knapsack Problem (QKP) involves selecting a subset of elements that maximizes the sum of pairwise and singleton utilities without exceeding a given budget. We introduce a Breakpoints Algorithm for QKP, named QKBP, which is based on a technique proposed in Hochbaum (2009) for efficiently generating the concave envelope of the solutions to the relaxation of the problem for all values of the budget. Our approach utilizes the fact that breakpoints in the concave envelopes are optimal solutions for their respective budgets. For budgets between breakpoints, a fast greedy procedure derives high-quality solutions from the optimal solutions of adjacent breakpoints. The algorithm is highly scalable due to an efficient parametric cut procedure used to generate the concave envelope. This efficiency is further improved by a newly developed compact problem formulation. Our extensive computational study on both existing and new benchmark instances, with up to 10,000 elements, shows that while some leading algorithms perform well on a few instances, QKBP consistently delivers high-quality solutions regardless of instance size, density, or budget. Moreover, QKBP achieves these results in significantly faster running times than all leading algorithms. The source code of the QKBP algorithm, the benchmark instances, and the detailed results are publicly available on GitHub.

The jet cross-section and jet-substructure observables in $p$$+$$p$ collisions at $\sqrt{s}=200$ GeV were measured by the PHENIX Collaboration at the Relativistic Heavy Ion Collider (RHIC). Jets are reconstructed from charged-particle tracks and electromagnetic-calorimeter clusters using the anti-$k_{t}$ algorithm with a jet radius $R=0.3$ for jets with transverse momentum within $8.0

Massless spinning correlators in cosmology are extremely complicated. In contrast, the scattering amplitudes of massless particles with spin are very simple. We propose that the reason for the unreasonable complexity of these correlators lies in the use of inconvenient kinematic variables. For example, in de Sitter space, consistency with unitarity and the background isometries imply that the correlators must be conformally covariant and also conserved. However, the commonly used kinematic variables for correlators do not make all of these properties manifest. In this paper, we introduce twistor space as a powerful way to satisfy all kinematic constraints. We show that conformal correlators of conserved currents can be written as twistor integrals, where the conservation condition translates into holomorphicity of the integrand. The functional form of the twistor-space correlators is very simple and easily bootstrapped. For the case of three-point functions, we verify explicitly that this reproduces known results in embedding space. We also perform a half-Fourier transform of the twistor-space correlators to obtain their counterparts in momentum space. We conclude that twistors provide a promising new avenue to study conformal correlation functions that exposes their hidden simplicity., Comment: Typo fixed and important references added

This study investigates the impact of varying sequence lengths on the accuracy of predicting cryptocurrency returns using Artificial Neural Networks (ANNs). Utilizing the Mean Absolute Error (MAE) as a threshold criterion, we aim to enhance prediction accuracy by excluding returns that are smaller than this threshold, thus mitigating errors associated with minor returns. The subsequent evaluation focuses on the accuracy of predicted returns that exceed this threshold. We compare four sequence lengths 168 hours (7 days), 72 hours (3 days), 24 hours, and 12 hours each with a return prediction interval of 2 hours. Our findings reveal the influence of sequence length on prediction accuracy and underscore the potential for optimized sequence configurations in financial forecasting models., Comment: 9 pages, 8 figures, 2 tables

Materials combining topologically non-trivial behavior and superconductivity offer a potential route for quantum computation. However, the set of available materials intrinsically realizing these properties are scarce. Recently, surface superconductivity has been reported in PtBi$_2$ in its trigonal phase and an inherent Weyl semimetal phase has been predicted. Here, based on scanning tunneling microscopy experiments, we reveal the signature of topological Fermi arcs in the normal state patterns of the quasiparticle interference. We show that the scattering between Fermi arcs dominates the interference spectra, providing conclusive evidence for the relevance of Weyl fermiology for the surface electronic properties of trigonal PtBi$_2$.

Learning how to safely handle Plasma-Material Interaction (PMI) is a key challenge towards the commercialisation of energy from nuclear fusion. In this respect, linear plasma devices are ideal experimental testbeds, and numerical codes play a crucial complementary role. In this paper, a numerical investigation of PMI-relevant helium plasma experimental discharges in GyM linear device is presented, in which SOLPS-ITER and ERO2.0 codes are coupled for plasma background generation and material erosion investigation respectively, with the aim to support the interpretation and complement the available experimental dataset. On the plasma side, simulated profiles are validated against experimental data to provide a realistic plasma background, and the role of He metastable states is assessed for the first time in SOLPS simulations. On the material side, the erosion and deposition effects due to the introduction of the sample-holder in the simulation volume are investigated, now considering also the real stainless steel composition as wall material.

This paper introduces an approach that integrates self-adaptive Evolution Strategies (ES) with Large Language Models (LLMs) to enhance the explainability of complex optimization processes. By employing a self-adaptive ES equipped with a restart mechanism, we effectively navigate the challenging landscapes of benchmark functions, capturing detailed logs of the optimization journey. The logs include fitness evolution, step-size adjustments and restart events due to stagnation. An LLM is then utilized to process these logs, generating concise, user-friendly summaries that highlight key aspects such as convergence behavior, optimal fitness achievements, and encounters with local optima. Our case study on the Rastrigin function demonstrates how our approach makes the complexities of ES optimization transparent. Our findings highlight the potential of using LLMs to bridge the gap between advanced optimization algorithms and their interpretability., Comment: Accepted at ESANN 2024

This paper we present our vision and ongoing work for a novel dataset designed to advance research into the interoperability of intelligent vehicles and infrastructure, specifically aimed at enhancing cooperative perception and interaction in the realm of public transportation. Unlike conventional datasets centered on ego-vehicle data, this approach encompasses both a stationary sensor tower and a moving vehicle, each equipped with cameras, LiDARs, and GNSS, while the vehicle additionally includes an inertial navigation system. Our setup features comprehensive calibration and time synchronization, ensuring seamless and accurate sensor data fusion crucial for studying complex, dynamic scenes. Emphasizing public transportation, the dataset targets to include scenes like bus station maneuvers and driving on dedicated bus lanes, reflecting the specifics of small public buses. We introduce the open-source ".4mse" file format for the new dataset, accompanied by a research kit. This kit provides tools such as ego-motion compensation or LiDAR-to-camera projection enabling advanced research on intelligent vehicle-infrastructure integration. Our approach does not include annotations; however, we plan to implement automatically generated labels sourced from state-of-the-art public repositories. Several aspects are still up for discussion, and timely feedback from the community would be greatly appreciated. A sneak preview on one data frame will be available at a Google Colab Notebook. Moreover, we will use the related GitHub Repository to collect remarks and suggestions.

The nonlinear interaction of x-rays with matter is at the heart of understanding and controlling ultrafast molecular dynamics from an atom-specific viewpoint, providing new scientific and analytical opportunities to explore the structure and dynamics of small quantum systems. At increasingly high x-ray intensity, the sensitivity of ultrashort x-ray pulses to specific electronic states and emerging short-lived transient intermediates is of particular relevance for our understanding of fundamental multi-photon absorption processes. In this work, intense x-ray free-electron laser (XFEL) pulses at the European XFEL (EuXFEL) are combined with a gas cell and grating spectrometer for a high-intensity transmission spectroscopy study of multiphoton-induced ultrafast molecular fragmentation dynamics in CF$_4$. This approach unlocks the direct intra-pulse observation of transient fragments, including neutral atoms, by their characteristic absorption lines in the transmitted broad-band x-ray spectrum. The dynamics with and without initially producing fluorine K-shell holes are studied by tuning the central photon energy. The absorption spectra are measured at different FEL intensities to observe nonlinear effects. Transient isolated fluorine atoms and ions are spectroscopically recorded within the ultrashort pulse duration of few tens of femtoseconds. An isosbestic point that signifies the correlated transition between intact neutral CF$_4$ molecules and charged atomic fragments is observed near the fluorine K-edge. The dissociation dynamics and the multiphoton absorption-induced dynamics encoded in the spectra are theoretically interpreted. Overall, this study demonstrates the potential of high-intensity x-ray transmission spectroscopy to study ultrafast molecular dynamics with sensitivity to specific intermediate species and their electronic structure., Comment: 30 pages, with 7 figures, submitted to Phys. Rev. X

Mechanical stresses and strains in the microstructure of cathode materials evolving during charge/discharge cycles can reduce the long-term stability of intercalation-type alkali-metal-ion batteries. In this context, crystalline compounds exhibiting zero-strain (ZS) behavior are of particular interest. Near-ZS sodiation was experimentally measured in the tetragonal tungsten bronze (TTB) type compound Na$_x$FeF$\mathrm{_3}$. Using a first-principles method based on density functional theory, we investigate the potential of iron-based fluoride compounds with tungsten bronze (TB) structures as ZS cathode materials. Simulations were conducted to study the intercalation of the alkali metal ions Li$\mathrm{^+}$, Na$\mathrm{^+}$, and K$\mathrm{^+}$ into the TTB and two related TB structures of the cubic perovskite (PTB) and hexagonal (HTB) types. We describe compensating local volume effects that can explain the experimentally measured low volume change of Na$_x$FeF$\mathrm{_3}$. We discuss the structural and chemical prerequisites of the host lattice for ZS insertion mechanism for alkali ions in TB structures and present a qualitative descriptor to predict the local volume change, that provides a way for faster screening and discovery of novel ZS battery materials., Comment: 10 pages, 6 figures

The technological properties of a system delivering simulation experience are a crucial dimension of immersion. To create a sense of presence and reproduce drivers behaviour as realistically as possible, we need reliable driving simulators that allow drivers to become highly immersed. This study investigates the impact of a system immersion of a driving simulator on the drivers' brain activity while operating a conditionally automated vehicle. Nineteen participants drove approximately 40 minutes while their brain activity was recorded using electroencephalography (EEG). We found a significant effect of the system immersion in the occipital and parietal areas, primarily in the high-Beta bandwidth. No effect was found in the Theta, Alpha, and low-Beta bandwidths. These findings suggest that the system immersion might influence the drivers' physiological arousal, consequently influencing their cognitive and emotional processes.

Accurately detecting dysfluencies in spoken language can help to improve the performance of automatic speech and language processing components and support the development of more inclusive speech and language technologies. Inspired by the recent trend towards the deployment of large language models (LLMs) as universal learners and processors of non-lexical inputs, such as audio and video, we approach the task of multi-label dysfluency detection as a language modeling problem. We present hypotheses candidates generated with an automatic speech recognition system and acoustic representations extracted from an audio encoder model to an LLM, and finetune the system to predict dysfluency labels on three datasets containing English and German stuttered speech. The experimental results show that our system effectively combines acoustic and lexical information and achieves competitive results on the multi-label stuttering detection task., Comment: Accepted at Interspeech 2024

This paper explores the improvement of post-training quantization (PTQ) after knowledge distillation in the Whisper speech foundation model family. We address the challenge of outliers in weights and activation tensors, known to impede quantization quality in transformer-based language and vision models. Extending this observation to Whisper, we demonstrate that these outliers are also present when transformer-based models are trained to perform automatic speech recognition, necessitating mitigation strategies for PTQ. We show that outliers can be reduced by a recently proposed gating mechanism in the attention blocks of the student model, enabling effective 8-bit quantization, and lower word error rates compared to student models without the gating mechanism in place., Comment: Accepted at Interspeech 2024

In this work, we optimize speculative sampling for parallel hardware accelerators to improve sampling speed. We notice that substantial portions of the intermediate matrices necessary for speculative sampling can be computed concurrently. This allows us to distribute the workload across multiple GPU threads, enabling simultaneous operations on matrix segments within thread blocks. This results in profiling time improvements ranging from 6% to 13% relative to the baseline implementation, without compromising accuracy. To further accelerate speculative sampling, probability distributions parameterized by softmax are approximated by sigmoid. This approximation approach results in significantly greater relative improvements in profiling time, ranging from 37% to 94%, with a minor decline in accuracy. We conduct extensive experiments on both automatic speech recognition and summarization tasks to validate the effectiveness of our optimization methods., Comment: Accepted at EMNLP 2024

Real-time bioaerosol monitoring is improving the quality of life for people affected by allergies, but it often relies on deep-learning models which pose challenges for widespread adoption. These models are typically trained in a supervised fashion and require considerable effort to produce large amounts of annotated data, an effort that must be repeated for new particles, geographical regions, or measurement systems. In this work, we show that self-supervised learning and few-shot learning can be combined to classify holographic images of bioaerosol particles using a large collection of unlabelled data and only a few examples for each particle type. We first demonstrate that self-supervision on pictures of unidentified particles from ambient air measurements enhances identification even when labelled data is abundant. Most importantly, it greatly improves few-shot classification when only a handful of labelled images are available. Our findings suggest that real-time bioaerosol monitoring workflows can be substantially optimized, and the effort required to adapt models for different situations considerably reduced., Comment: Short communication, 8 pages, 2 figures, 1 table

Automated vehicles are pictured as the future of transportation, and facilitating safer driving is only one of the many benefits. However, due to the constantly changing role of the human driver, users are easily confused and have little knowledge about their responsibilities. Being the bridge between automation and human, the human-machine interface (HMI) is of great importance to driving safety. This study was conducted in a static driving simulator. Three HMI designs were developed, among which significant differences in mental workload using NASA-TLX and the subjective transparency test were found. An electroencephalogram was applied throughout the study to determine if differences in the mental workload could also be found using EEG's spectral power analysis. Results suggested that more studies are required to determine the effectiveness of the spectral power of EEG on mental workload, but the three interface designs developed in this study could serve as a solid basis for future research to evaluate the effectiveness of psychophysiological measures. Marie Sklodowska-Curie Actions; Innovative Training Network (ITN); SHAPE-IT; Grant number 860410; Publication date: [27 July 2023]; DOI: [10.1109/IV55152.2023.10186567]

Human-machine Interface (HMI) is critical for safety during automated driving, as it serves as the only media between the automated system and human users. To enable a transparent HMI, we first need to know how to evaluate it. However, most of the assessment methods used for HMI designs are subjective and thus not efficient. To bridge the gap, an objective and standardized HMI assessment method is needed, and the first step is to find an objective method for workload measurement for this context. In this study, two psychophysiological measures, electrocardiography (ECG) and electrodermal activity (EDA), were evaluated for their effectiveness in finding differences in mental workload among different HMI designs in a simulator study. Three HMI designs were developed and used. Results showed that both workload measures were able to identify significant differences in objective mental workload when interacting with in-vehicle HMIs. As a first step toward a standardized assessment method, the results could be used as a firm ground for future studies. Marie Sk{\l}odowska-Curie Actions; Innovative Training Network (ITN); SHAPE-IT; Grant number 860410; Publication date: [29 Sep 2023]; DOI: [10.54941/ahfe1004172]

High-momentum two-particle correlations are a useful tool for studying jet-quenching effects in the quark-gluon plasma. Angular correlations between neutral-pion triggers and charged hadrons with transverse momenta in the range 4--12~GeV/$c$ and 0.5--7~GeV/$c$, respectively, have been measured by the PHENIX experiment in 2014 for Au$+$Au collisions at $\sqrt{s_{_{NN}}}=200$~GeV. Suppression is observed in the yield of high-momentum jet fragments opposite the trigger particle, which indicates jet suppression stemming from in-medium partonic energy loss, while enhancement is observed for low-momentum particles. The ratio and differences between the yield in Au$+$Au collisions and $p$$+$$p$ collisions, $I_{AA}$ and $\Delta_{AA}$, as a function of the trigger-hadron azimuthal separation, $\Delta\phi$, are measured for the first time at the Relativistic Heavy Ion Collider. These results better quantify how the yield of low-$p_T$ associated hadrons is enhanced at wide angle, which is crucial for studying energy loss as well as medium-response effects., Comment: 535 authors from 84 institutions, 12 pages, 8 figures. v2 is version accepted for publication in Physical Review C. HEPdata tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html

The omega-regular separability problem for B\"uchi VASS coverability languages has recently been shown to be decidable, but with an EXPSPACE lower and a non-primitive recursive upper bound -- the exact complexity remained open. We close this gap and show that the problem is EXPSPACE-complete. A careful analysis of our complexity bounds additionally yields a PSPACE procedure in the case of fixed dimension >= 1, which matches a pre-established lower bound of PSPACE for one dimensional B\"uchi VASS. Our algorithm is a non-deterministic search for a witness whose size, as we show, can be suitably bounded. Part of the procedure is to decide the existence of runs in VASS that satisfy certain non-linear properties. Therefore, a key technical ingredient is to analyze a class of systems of inequalities where one variable may occur in non-linear (polynomial) expressions. These so-called singly non-linear systems (SNLS) take the form A(x).y >= b(x), where A(x) and b(x) are a matrix resp. a vector whose entries are polynomials in x, and y ranges over vectors in the rationals. Our main contribution on SNLS is an exponential upper bound on the size of rational solutions to singly non-linear systems. The proof consists of three steps. First, we give a tailor-made quantifier elimination to characterize all real solutions to x. Second, using the root separation theorem about the distance of real roots of polynomials, we show that if a rational solution exists, then there is one with at most polynomially many bits. Third, we insert the solution for x into the SNLS, making it linear and allowing us to invoke standard solution bounds from convex geometry. Finally, we combine the results about SNLS with several techniques from the area of VASS to devise an EXPSPACE decision procedure for omega-regular separability of B\"uchi VASS.

Drivers' role changes with increasing automation from the primary driver to a system supervisor. This study investigates how supervising an SAE L2 and L3 automated vehicle (AV) affects drivers' mental workload and sleepiness compared to manual driving. Using an AV prototype on a test track, the oscillatory brain activity of 23 adult participants was recorded during L2, L3, and manual driving. Results showed decreased mental workload and increased sleepiness in L3 drives compared to L2 and manual drives, indicated by self-report scales and changes in the frontal alpha and theta power spectral density. These findings suggest that fatigue and mental underload are significant issues in L3 driving and should be considered when designing future AV interfaces.

Manipulating MW in driving simulator studies without the need to introduce a non-driving-related task remains challenging. This study aims to empirically evaluate the modified speed regulation n-back task, a tool to manipulate drivers' MW. Our experiment involved 23 participants who experienced a 0-back and 2-back driving condition, with task-irrelevant novel environmental sounds used to elicit P3a event-related potentials. Results indicate that the 2-back condition was perceived as more demanding, evidenced by higher NASA-TLX scores (overall score, mental and temporal demand, effort, frustration). The mean P3a amplitude was diminished during the 2-back condition compared to the 0-back condition, suggesting that drivers experienced higher MW and had fewer resources available to process the novel environmental sounds. This study provides empirical evidence indicating that the speed regulation n-back task could be a valid, effective, and reproducible method to manipulate MW in driving research.

Improved understandings of two-phase transport in electrochemical gas-evolving systems are increasingly demanded, while high-performance imaging techniques using simplified instrumentations are not readily available. This work presents volumetric scanning acoustic microscopy (SAM) imaging for quantifying the dynamics of gas bubbles and electrolyte in porous Nickel electrodes with different wettability and structures during alkaline water electrolysis (AWE). We realize high-resolution 3D imaging at 10's um level using high frequency spherically focused ultrasound. The high resolution allowed us to clearly visualize the spatial distributions of produced bubbles in the porous electrodes over time. Moreover, we are able to quantify the residual gas volume in an electrode and its coverage due to bubble evolution, which dominate its transport overpotential. Taking these advantages, we elucidate the impacts of electrodes' wettability and structures on their electrolysis performance, on a regular laboratory base. The obtained knowledge provides us important optimization guidelines of AWE designs and operating schemes., Comment: Research artical on an emerging field. 33 pages, 6 figures, 61 references, 10 supplementary figures available. Journal submission in progress

Traditional recommendation proposals, including content-based and collaborative filtering, usually focus on similarity between items or users. Existing approaches lack ways of introducing unexpectedness into recommendations, prioritizing globally popular items over exposing users to unforeseen items. This investigation aims to design and evaluate a novel layer on top of recommender systems suited to incorporate relational information and suggest items with a user-defined degree of surprise. We propose a Knowledge Graph (KG) based recommender system by encoding user interactions on item catalogs. Our study explores whether network-level metrics on KGs can influence the degree of surprise in recommendations. We hypothesize that surprisingness correlates with certain network metrics, treating user profiles as subgraphs within a larger catalog KG. The achieved solution reranks recommendations based on their impact on structural graph metrics. Our research contributes to optimizing recommendations to reflect the metrics. We experimentally evaluate our approach on two datasets of LastFM listening histories and synthetic Netflix viewing profiles. We find that reranking items based on complex network metrics leads to a more unexpected and surprising composition of recommendation lists.

Text-to-image generative models have become a prominent and powerful tool that excels at generating high-resolution realistic images. However, guiding the generative process of these models to consider detailed forms of conditioning reflecting style and/or structure information remains an open problem. In this paper, we present LoRAdapter, an approach that unifies both style and structure conditioning under the same formulation using a novel conditional LoRA block that enables zero-shot control. LoRAdapter is an efficient, powerful, and architecture-agnostic approach to condition text-to-image diffusion models, which enables fine-grained control conditioning during generation and outperforms recent state-of-the-art approaches., Comment: for the project page and code, view https://compvis.github.io/LoRAdapter/

Neural network representations of simple models, such as linear regression, are being studied increasingly to better understand the underlying principles of deep learning algorithms. However, neural representations of distributional regression models, such as the Cox model, have received little attention so far. We close this gap by proposing a framework for distributional regression using inverse flow transformations (DRIFT), which includes neural representations of the aforementioned models. We empirically demonstrate that the neural representations of models in DRIFT can serve as a substitute for their classical statistical counterparts in several applications involving continuous, ordered, time-series, and survival outcomes. We confirm that models in DRIFT empirically match the performance of several statistical methods in terms of estimation of partial effects, prediction, and aleatoric uncertainty quantification. DRIFT covers both interpretable statistical models and flexible neural networks opening up new avenues in both statistical modeling and deep learning., Comment: Accepted at UAI 2024 https://www.auai.org/uai2024/accepted_papers

Context: Cultural aspects are of high importance as they guide people's behaviour and thus, influence how people apply methods and act in projects. In recent years, software engineering research emphasized the need to analyze the challenges of specific cultural characteristics. Investigating the influence of cultural characteristics is challenging due to the multi-faceted concept of culture. People's behaviour, their beliefs and underlying values are shaped by different layers of culture, e.g., regions, organizations, or groups. In this study, we focus on agile methods, which are agile approaches that focus on underlying values, collaboration and communication. Thus, cultural and social aspects are of high importance for their successful use in practice. Objective: In this paper, we address challenges that arise when using the model of cultural dimensions by Hofstede to characterize specific cultural values. This model is often used when discussing cultural influences in software engineering. Method: As a basis, we conducted an exploratory, multiple case study, consisting of two cases in Japan and two in Germany. Contributions: In this study, we observed that cultural characteristics of the participants differed significantly from cultural characteristics that would typically be expected for people from the respective country. This drives our conclusion that for studies in empirical software engineering that address cultural factors, a case-specific analysis of the characteristics is needed.

Contemporary accelerator designs exhibit a high degree of spatial localization, wherein two-dimensional physical distance determines communication costs between processing elements. This situation presents considerable algorithmic challenges, particularly when managing sparse data, a pivotal component in progressing data science. The spatial computer model quantifies communication locality by weighting processor communication costs by distance, introducing a term named energy. Moreover, it integrates depth, a widely-utilized metric, to promote high parallelism. We propose and analyze a framework for efficient spatial tree algorithms within the spatial computer model. Our primary method constructs a spatial tree layout that optimizes the locality of the neighbors in the compute grid. This approach thereby enables locality-optimized messaging within the tree. Our layout achieves a polynomial factor improvement in energy compared to utilizing a PRAM approach. Using this layout, we develop energy-efficient treefix sum and lowest common ancestor algorithms, which are both fundamental building blocks for other graph algorithms. With high probability, our algorithms exhibit near-linear energy and poly-logarithmic depth. Our contributions augment a growing body of work demonstrating that computations can have both high spatial locality and low depth. Moreover, our work constitutes an advancement in the spatial layout of irregular and sparse computations.

The Industrial Internet of Things (IIoT) is reshaping manufacturing, industrial processes, and infrastructure management. By fostering new levels of automation, efficiency, and predictive maintenance, IIoT is transforming traditional industries into intelligent, seamlessly interconnected ecosystems. However, achieving highly reliable IIoT can be hindered by factors such as the cost of installing large numbers of sensors, limitations in retrofitting existing systems with sensors, or harsh environmental conditions that may make sensor installation impractical. Soft (virtual) sensing leverages mathematical models to estimate variables from physical sensor data, offering a solution to these challenges. Data-driven and physics-based modeling are the two main methodologies widely used for soft sensing. The choice between these strategies depends on the complexity of the underlying system, with the data-driven approach often being preferred when the physics-based inference models are intricate and present challenges for state estimation. However, conventional deep learning models are typically hindered by their inability to explicitly represent the complex interactions among various sensors. To address this limitation, we adopt Graph Neural Networks (GNNs), renowned for their ability to effectively capture the complex relationships between sensor measurements. In this research, we propose physics-enhanced GNNs, which integrate principles of physics into graph-based methodologies. This is achieved by augmenting additional nodes in the input graph derived from the underlying characteristics of the physical processes. Our evaluation of the proposed methodology on the case study of district heating networks reveals significant improvements over purely data-driven GNNs, even in the presence of noise and parameter inaccuracies., Comment: 14 pages, 10 figures. Accepted to IEEE Internet of Things Journal

The need for compliant and proprioceptive actuators has grown more evident in pursuing more adaptable and versatile robotic systems. Hydraulically Amplified Self-Healing Electrostatic (HASEL) actuators offer distinctive advantages with their inherent softness and flexibility, making them promising candidates for various robotic tasks, including delicate interactions with humans and animals, biomimetic locomotion, prosthetics, and exoskeletons. This has resulted in a growing interest in the capacitive self-sensing capabilities of HASEL actuators to create miniature displacement estimation circuitry that does not require external sensors. However, achieving HASEL self-sensing for actuation frequencies above 1 Hz and with miniature high-voltage power supplies has remained limited. In this paper, we introduce the F-HASEL actuator, which adds an additional electrode pair used exclusively for capacitive sensing to a Peano-HASEL actuator. We demonstrate displacement estimation of the F-HASEL during high-frequency actuation up to 20 Hz and during external loading using miniaturized circuitry comprised of low-cost off-the-shelf components and a miniature high-voltage power supply. Finally, we propose a circuitry to estimate the displacement of multiple F-HASELs and demonstrate it in a wearable application to track joint rotations of a virtual reality user in real-time., Comment: This work has been submitted to the IEEE for possible publication