Search

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 3,000 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai., Comment: 25 pages, 6 figures

We present an all-silicon source of near-infrared linearly-polarized single photons, fabricated by nanoscale positioning of a color center in a silicon-on-insulator microcavity. The color center consists of a single W center, created at a well-defined position by Si$^{+}$ ion implantation through a 150 nm-diameter nanohole in a mask. A circular Bragg grating cavity resonant with the W's zero-phonon line at 1217 nm is fabricated at the same location as the nanohole. Under above-gap continuous-wave excitation, a very clean photon antibunching behavior ($g{^2} \leq 0.06$) is observed over the entire power range, which highlights the absence of parasitic emitters. Purcell-enhancement of W's zero-phonon emission provides both a record-high photoluminescence count rate among Si color centers (ca $1.2 \times 10^{6}$ counts/s) and apparent Debye-Waller factor around 99%. We also demonstrate the triggered emission of single photons with 93% purity under weak pulsed laser excitation. At high pulsed laser power, we reveal a detrimental effect of repumping processes, that could be mitigated using selective pumping schemes in the future. These results represent a major step towards on-demand sources of indistinguishable near-infrared single photons within silicon photonics chips.

The average state dynamics of free fermions subject to random projective measurements of local site occupation numbers is governed by a Lindblad equation with dephasing noise. In the continuous limit, the equation of motion for the correlation matrix is mapped to a hydrodynamic equation for the Wigner function, which takes the form of a linear, nonlocal partial differential equation. As a main result, we derive the analytical solution to the hydrodynamic equation using the quasiparticle picture, showing that the Wigner dynamics emerges from stochastic sampling of classical run-and-tumble processes. As an application, we recover the crossover between ballistic and diffusive transport regimes., Comment: 9 pages, 3 figures

Two abelian varieties $A$ and $B$ over a number field $K$ are said to be strongly locally quadratic twists if they are quadratic twists at every completion of $K$. While it was known that this does not imply that $A$ and $B$ are quadratic twists over $K$, the only known counterexamples (necessarily of dimension $\geq 4$) are not geometrically simple. We show that, for every prime $p\equiv 13 \pmod{24}$, there exists a pair of geometrically simple abelian varieties of dimension $p-1$ over $\mathbb{Q}$ that are strongly locally quadratic twists but not quadratic twists. The proof is based on Galois cohomology computations and class field theory., Comment: 15 pages. Comments are welcome!

Magnetic fields are ubiquitous across different physical systems of current interest; from the early Universe, compact astrophysical objects and heavy-ion collisions to condensed matter systems. A proper treatment of the effects produced by magnetic fields during the dynamical evolution of these systems, can help to understand observables that otherwise show a puzzling behavior. Furthermore, when these fields are comparable to or stronger than \Lambda_QCD, they serve as excellent probes to help elucidate the physics of strongly interacting matter under extreme conditions of temperature and density. In this work we provide a comprehensive review of recent developments on the description of QED and QCD systems where magnetic field driven effects are important. These include the modification of meson static properties such as masses and form factors, the chiral magnetic effect, the description of anomalous transport coefficients, superconductivity in extreme magnetic fields, the properties of neutron stars, the evolution of heavy-ion collisions, as well as effects on the QCD phase diagram. We describe recent theory and phenomenological developments using effective models as well as LQCD methods. The work represents a state-of-the-art review of the field, motivated by presentations and discussions during the "Workshop on Strongly Interacting Matter in Strong Electromagnetic Fields" that took place in the European Centre for Theoretical Studies in Nuclear Physics and Related Areas (ECT*) in the city of Trento, Italy, September 25-29, 2023., Comment: 325 pages-long review of recent topics of interest in the field of magnetic field effects on QED and QCD matter. To be susbmitted to PNPP

Dynamical phase transitions induced by local projective measurements have attracted a lot of attention in the past few years. It has been in particular argued that measurements may induce an abrupt change in the scaling law of the bipartite entanglement entropy. In this work we show that local projective measurements on a one-dimensional quadratic fermionic system induce a qualitative modification of the time growth of the entanglement entropy, changing from linear to logarithmic. However, in the stationary regime, the logarithmic behavior of the entanglement entropy does not survive in the thermodynamic limit and, for any finite value of the measurement rate, we numerically show the existence of a single area-law phase for the entanglement entropy. We give analytical arguments supporting our conclusions., Comment: 7 pages, 4 figures

This study investigates the temperature-induced hysteretic behavior of resistivity and magnetoresistance in electrodeposited bismuth films, with a focus on their application as absorbers in transition-edge sensors (TESs) for X-ray detection. Through a series of resistance versus temperature measurements from room temperature to a few Kelvins, we explore the change in the conductive behavior of bismuth electrodeposited on various substrates at the various temperatures. Our findings show for the first time both hysteretic and irreversible changes in resistivity as a function of temperature. Further, magnetoresistance measurements reveal notable variations in resistance behavior under different magnetic fields, highlighting the impact of magnetic fields on these films' electronic transport properties, with an indication of potential weak anti-localization effects at the lowest temperatures. This study not only provides a deeper understanding of bismuth's conductivity characteristics at low temperatures but also sheds light on the practical implications for developing more effective TESs for synchrotron X-ray facilities.

Dexterous robotic manipulator teleoperation is widely used in many applications, either where it is convenient to keep the human inside the control loop, or to train advanced robot agents. So far, this technology has been used in combination with camera systems with remarkable success. On the other hand, only a limited number of studies have focused on leveraging haptic feedback from tactile sensors in contexts where camera-based systems fail, such as due to self-occlusions or poor light conditions like smoke. This study demonstrates the feasibility of precise pick-and-place teleoperation without cameras by leveraging tactile-based 3D object reconstruction in VR and providing haptic feedback to a blindfolded user. Our preliminary results show that integrating these technologies enables the successful completion of telemanipulation tasks previously dependent on cameras, paving the way for more complex future applications., Comment: 7 pages, 5 figures, This work has been submitted to the IEEE for possible publication

Robotic telemanipulation - the human-guided manipulation of remote objects - plays a pivotal role in several applications, from healthcare to operations in harsh environments. While visual feedback from cameras can provide valuable information to the human operator, haptic feedback is essential for accessing specific object properties that are difficult to be perceived by vision, such as stiffness. For the first time, we present a participant study demonstrating that operators can perceive the stiffness of remote objects during real-world telemanipulation with a dexterous robotic hand, when haptic feedback is generated from tactile sensing fingertips. Participants were tasked with squeezing soft objects by teleoperating a robotic hand, using two methods of haptic feedback: one based solely on the measured contact force, while the second also includes the squeezing displacement between the leader and follower devices. Our results demonstrate that operators are indeed capable of discriminating objects of different stiffness, relying on haptic feedback alone and without any visual feedback. Additionally, our findings suggest that the displacement feedback component may enhance discrimination with objects of similar stiffness., Comment: 8 pages,6 figures, "This work has been submitted to the IEEE for possible publication."

This paper introduces a family of entropy-conserving finite-difference discretizations for the compressible flow equations. In addition to conserving the primary quantities of mass, momentum, and total energy, the methods also preserve kinetic energy and pressure equilibrium. The schemes are based on finite-difference (FD) representations of the logarithmic mean, establishing and leveraging a broader link between linear and nonlinear two-point averages and FD forms. The schemes are locally conservative due to the summation-by-parts property and therefore admit a local flux form, making them applicable also in finite-volume and finite-element settings. The effectiveness of these schemes is validated through various test cases (1D Sod shock tube, 1D density wave, 2D isentropic vortex, 3D Taylor Green vortex) that demonstrate exact conservation of entropy along with conservation of the primary quantities and preservation of pressure equilibrium., Comment: 25 pages, 7 figures

Adaptive gradient methods have been increasingly adopted by deep learning community due to their fast convergence and reduced sensitivity to hyper-parameters. However, these methods come with limitations, such as increased memory requirements for elements like moving averages and a poorly understood convergence theory. To overcome these challenges, we introduce F-CMA, a Fast-Controlled Mini-batch Algorithm with a random reshuffling method featuring a sufficient decrease condition and a line-search procedure to ensure loss reduction per epoch, along with its deterministic proof of global convergence to a stationary point. To evaluate the F-CMA, we integrate it into conventional training protocols for classification tasks involving both convolutional neural networks and vision transformer models, allowing for a direct comparison with popular optimizers. Computational tests show significant improvements, including a decrease in the overall training time by up to 68%, an increase in per-epoch efficiency by up to 20%, and in model accuracy by up to 5%., Comment: There is an error in the literature review, in section 1. In particular, we noticed that there is a wrong citation, the [65], which has been erroneously associated with another author's claims

We calculate the screening masses of pseudoscalar mesons in a hot and strongly magnetized medium within the framework of the SU(3) Nambu-Jona--Lasinio model, using a magnetic field-independent regularization scheme. Inverse magnetic catalysis (IMC) is implemented through the use of a magnetic field-dependent coupling $G(B)$, fitted to reproduce lattice QCD results for the pseudocritical chiral transition temperature $T_c^B$. For the external homogeneous magnetic field considered, neutral screening masses separate in two types: perpendicular and parallel to the direction of the field, while for charged mesons only parallel energies can be defined for each Landau level. We obtain $m_{\mathrm{scr},\perp} > m_{\mathrm{scr},\parallel}$, as expected from causality. Thermally, all screening energies are almost constant until some critical temperature, whose behavior is correlated with $T_c^B$. They rapidly increase around this value, keeping a steady enhancement afterwards due to thermal excitation. Magnetically, neutral parallel masses are enhanced (suppressed) at high temperatures when considering $G(B)$ ($G$). Perpendicular ones display a non-linear magnetic behavior for $G$ (due to increasing $T_c^B$), while they always increase with $B$ for $G(B)$. Charged parallel energies are also always magnetically enhanced. In the high-temperature limit, we show that both neutral and charged screening energies converge to $2\pi T$. At $B=0$ the model overestimates the remaining quark interaction in this regime. At $B\neq 0$ we find that, when IMC is accounted for, the interaction is suppressed as $B$ increases, a fact that appears to be at odds with currently available lattice QCD results., Comment: 38 pages, 11 figures

The effects of kinetic-energy preservation errors due to Runge-Kutta (RK) temporal integrators have been analyzed for the case of large-eddy simulations of incompressible turbulent channel flow. Simulations have been run using the open-source solver Xcompact3D with an implicit spectral vanishing viscosity model and a variety of temporal Runge-Kutta integrators. Explicit pseudo-symplectic schemes, with improved energy preservation properties, have been compared to standard RK methods. The results show a marked decrease in the temporal error for higher-order pseudo-symplectic methods; on the other hand, an analysis of the energy spectra indicates that the dissipation introduced by the commonly used three-stage RK scheme can lead to significant distortion of the energy distribution within the inertial range. A cost-vs-accuracy analysis suggests that pseudo-symplectic schemes could be used to attain results comparable to traditional methods at a reduced computational cost., Comment: 13 pages, 11 figures. F.C. and G.C. contributed equally to this work

This study proposes a novel spatial discretization procedure for the compressible Euler equations which guarantees entropy conservation at a discrete level when an arbitrary equation of state is assumed. The proposed method, based on a locally-conservative discretization, guarantees also the spatial conservation of mass, momentum, and total energy and is kinetic energy-preserving. In order to achieve the entropy-conservation property for an arbitrary non-ideal gas, a general strategy is adopted based on the manipulation of discrete balance equations through the imposition of global entropy conservation and the use of a summation by parts rule. The procedure, which is extended to an arbitrary order of accuracy, conducts to a general form of the internal-energy numerical flux which results in a nonlinear function of thermodynamic and dynamic variables and still admits the mass flux as a residual degree of freedom. The effectiveness of the novel entropy-conservative formulation is demonstrated through numerical tests making use of some of the most popular cubic equations of state., Comment: 16 pages, 3 figures

We study in detail the Ramanujan smooth expansions, for arithmetic functions; we start with the most general ones, for which we supply the "$P-$local expansions", for arguments with all prime-factors $p\le P$ (namely, $P-$smooth arguments), that are also square-free; then, we supply general results for interesting subsets of arithmetic functions, regarding both their $P-$local and (global) Ramanujan smooth expansions., Comment: Plain TeX, 20 pages, plus 1 p.bib

Six years after the entry into force of the GDPR, European companies and organizations still have difficulties complying with it: the amount of fines issued by the European data protection authorities is continuously increasing. Personal data transfers are no exception. In this work we analyse the personal data transfers from more than 20000 Italian Public Administration (PA) entities to third countries. We developed "Minos", a user-friendly application which allows to navigate the web while recording HTTP requests. Then, we used the back-end of Minos to automate the analysis. We found that about 14% of the PAs websites transferred data out of the European Economic Area (EEA). This number is an underestimation because only visits to the home pages were object of the analysis. The top 3 destinations of the data transfers are Amazon, Google and Fonticons, accounting for about the 70% of the bad requests. The most recurrent services which are the object of the requests are cloud computing services and content delivery networks (CDNs). Our results highlight that, in Italy, a relevant portion of public administrations websites transfers personal data to non EEA countries. In terms of technology policy, these results stress the need for further incentives to improve the PA digital infrastructures. Finally, while working on refinements of Minos, the version here described is openly available on Zenodo: it can be helpful to a variety of actors (citizens, researchers, activists, policy makers) to increase awareness and enlarge the investigation., Comment: International Conference on Information Technology for Social Good (GoodIT '24), September 4-6, 2024, Bremen, Germany

Employing physically-consistent numerical methods is an important step towards attaining robust and accurate numerical simulations. When addressing compressible flows, in addition to preserving kinetic energy at a discrete level, as done in the incompressible case, additional properties are sought after, such as the ability to preserve the equilibrium of pressure that can be found at contact interfaces. This paper investigates the general conditions of the spatial numerical discretizations to achieve the pressure equilibrium preserving property (PEP). Schemes from the literature are analyzed in this respect, and procedures to impart the PEP property to existing discretizations are proposed. Additionally, new PEP numerical schemes are introduced through minor modifications of classical ones. Numerical tests confirmed the theory hereby presented and showed that the modifications, beyond the enforcement of the PEP property, have a generally positive impact on the performances of the original schemes., Comment: 15 pages, 2 figures

This study explores the impact of adversarial perturbations on Convolutional Neural Networks (CNNs) with the aim of enhancing the understanding of their underlying mechanisms. Despite numerous defense methods proposed in the literature, there is still an incomplete understanding of this phenomenon. Instead of treating the entire model as vulnerable, we propose that specific feature maps learned during training contribute to the overall vulnerability. To investigate how the hidden representations learned by a CNN affect its vulnerability, we introduce the Adversarial Intervention framework. Experiments were conducted on models trained on three well-known computer vision datasets, subjecting them to attacks of different nature. Our focus centers on the effects that adversarial perturbations to a model's initial layer have on the overall behavior of the model. Empirical results revealed compelling insights: a) perturbing selected channel combinations in shallow layers causes significant disruptions; b) the channel combinations most responsible for the disruptions are common among different types of attacks; c) despite shared vulnerable combinations of channels, different attacks affect hidden representations with varying magnitudes; d) there exists a positive correlation between a kernel's magnitude and its vulnerability. In conclusion, this work introduces a novel framework to study the vulnerability of a CNN model to adversarial perturbations, revealing insights that contribute to a deeper understanding of the phenomenon. The identified properties pave the way for the development of efficient ad-hoc defense mechanisms in future applications., Comment: 22 pages, 15 figures (including appendix)

Recently, the n$^3$He collaboration reported a measurement of the parity-violating (PV) proton directional asymmetry $A_{\mathrm {PV}} = (1.55\pm 0.97~\mathrm {(st\ at)} \pm 0.24~\mathrm {(sys)})\times 10^{-8}$ in the capture reaction of ${}^3$He$(\vec {n},{\mathrm p}){}^3$H at meV incident neutron energies. The result increased the limited inventory of precisely measured and calculable PV observables in few-body systems required to further understand the structure of hadronic weak interaction. In this letter, we report the experimental and theoretical investigation of a parity conserving (PC) asymmetry $A_{\mathrm {PC}}$ in the same reaction (the first ever measured PC observable at meV neutron energies). As a result of S- and P-wave mixing in the reaction, the $A_{\mathrm {PC}}$ is inversely proportional to the neutron wavelength $\lambda$. The experimental value is $(\lambda\times A_{\mathrm {PC}})\equiv\beta= (-1.97 \pm 0.28~\mathrm{(stat)}\pm 0.12~\mathrm{(sys)}) \times 10^{-6}$ Amstrongs. We present results for a theoretical analysis of this reaction by solving the four-body scattering problem within the hyperspherical harmonic method. We find that in the ${}^3$He$(\vec {n},{\mathrm p}){}^3$H reaction, $A_{\mathrm {PC}}$ depends critically on the energy and width of the close $0^-$ resonant state of ${}^4$He, resulting in a large sensitivity to the spin-orbit components of the nucleon-nucleon force and even to the three-nucleon force. The analysis of the accurately measured $A_{\mathrm {PC}}$ and $A_{\mathrm {PV}}$ using the same few-body theoretical models gives essential information needed to interpret the PV asymmetry in the ${}^3$He$(\vec {n}, {\mathrm p}){}^3$H reaction., Comment: 8 pages, 3 figures

Large Language Models (LLMs) are becoming key in automating and assisting various software development tasks, including text-based tasks in requirements engineering but also in coding. Typically, these models are used to automate small portions of existing tasks, but we present a broader vision to span multiple steps from requirements engineering to implementation using existing libraries. This approach, which we call Semantic API Alignment (SEAL), aims to bridge the gap between a user's high-level goals and the specific functions of one or more APIs. In this position paper, we propose a system architecture where a set of LLM-powered ``agents'' match such high-level objectives with appropriate API calls. This system could facilitate automated programming by finding matching links or, alternatively, explaining mismatches to guide manual intervention or further development. As an initial pilot, our paper demonstrates this concept by applying LLMs to Goal-Oriented Requirements Engineering (GORE), via sub-goal analysis, for aligning with REST API specifications, specifically through a case study involving a GitHub statistics API. We discuss the potential of our approach to enhance complex tasks in software development and requirements engineering and outline future directions for research.

Effective execution of long-horizon tasks with dexterous robotic hands remains a significant challenge in real-world problems. While learning from human demonstrations have shown encouraging results, they require extensive data collection for training. Hence, decomposing long-horizon tasks into reusable primitive skills is a more efficient approach. To achieve so, we developed DexSkills, a novel supervised learning framework that addresses long-horizon dexterous manipulation tasks using primitive skills. DexSkills is trained to recognize and replicate a select set of skills using human demonstration data, which can then segment a demonstrated long-horizon dexterous manipulation task into a sequence of primitive skills to achieve one-shot execution by the robot directly. Significantly, DexSkills operates solely on proprioceptive and tactile data, i.e., haptic data. Our real-world robotic experiments show that DexSkills can accurately segment skills, thereby enabling autonomous robot execution of a diverse range of tasks.

The paper aims to investigate relevant computational issues of deep neural network architectures with an eye to the interaction between the optimization algorithm and the classification performance. In particular, we aim to analyze the behaviour of state-of-the-art optimization algorithms in relationship to their hyperparameters setting in order to detect robustness with respect to the choice of a certain starting point in ending on different local solutions. We conduct extensive computational experiments using nine open-source optimization algorithms to train deep Convolutional Neural Network architectures on an image multi-class classification task. Precisely, we consider several architectures by changing the number of layers and neurons per layer, in order to evaluate the impact of different width and depth structures on the computational optimization performance.

In autonomous mobility-on-demand systems, effectively managing vehicle flows to mitigate induced congestion and ensure efficient operations is imperative for system performance and positive customer experience. Against this background, we study the potential of staggered routing, i.e., purposely delaying trip departures from a system perspective, in order to reduce congestion and ensure efficient operations while still meeting customer time windows. We formalize the underlying planning problem and show how to efficiently model it as a mixed integer linear program. Moreover, we present a matheuristic that allows us to efficiently solve large-scale real-world instances both in an offline full-information setting and its online rolling horizon counterpart. We conduct a numerical study for Manhattan, New York City, focusing on low- and highly-congested scenarios. Our results show that in low-congestion scenarios, staggering trip departures allows mitigating, on average, 94% of the induced congestion in a full information setting. In a rolling horizon setting, our algorithm allows us to reduce 90% of the induced congestion. In high-congestion scenarios, we observe an average reduction of 66% as the full information bound and an average reduction of 56% in our online setting. Surprisingly, we show that these reductions can be reached by shifting trip departures by a maximum of six minutes in both the low and high-congestion scenarios., Comment: 32 pages, 10 figures

The Scrum framework has gained widespread adoption in the industry for its emphasis on collaboration and continuous improvement. However, it has not reached a similar relevance in Software Engineering (SE) curricula. This work reports the experience of five editions of a SE course within an MSc. Degree in Computer Engineering. The course primary educational objective is to provide students with the skills to manage software development projects with Scrum. The course is based on the execution of a team project and on the definition of qualitative and quantitative means of assessment of the application of Scrum. The conduction of five editions of the course allowed us to identify several lessons learned about time budgeting and team compositions in agile student projects and its evidence of the applicability of the framework to software development courses., Comment: Submitted to LEARNER 2024 workshop

The abstraction of dynamical systems is a powerful tool that enables the design of feedback controllers using a correct-by-design framework. We investigate a novel scheme to obtain data-driven abstractions of discrete-time stochastic processes in terms of richer discrete stochastic models, whose actions lead to nondeterministic transitions over the space of probability measures. The data-driven component of the proposed methodology lies in the fact that we only assume samples from an unknown probability distribution. We also rely on the model of the underlying dynamics to build our abstraction through backward reachability computations. The nondeterminism in the probability space is captured by a collection of Markov Processes, and we identify how this model can improve upon existing abstraction techniques in terms of satisfying temporal properties, such as safety or reach-avoid. The connection between the discrete and the underlying dynamics is made formal through the use of the scenario approach theory. Numerical experiments illustrate the advantages and main limitations of the proposed techniques with respect to existing approaches., Comment: 6 pages

Software testing is a crucial phase in software development, enabling the detection of issues and defects that may arise during the development process. Addressing these issues enhances software applications' quality, reliability, user experience, and performance. Graphical User Interface (GUI) testing, one such technique, involves mimicking a regular user's interactions with an application to identify defects. However, GUI testing is often underutilized due to its perceived repetitiveness, error-proneness, and lack of immediate feedback on test quality. In recent years, gamification-incorporating game elements in non-game contexts to boost interest, motivation, and engagement-has gained traction in various fields, including software engineering and education. This paper presents GIPGUT: a prototype of a gamification plugin for IntelliJ IDEA, an Integrated Development Environment (IDE) that supports scripted GUI testing. The plugin enhances testers' engagement with typically monotonous and tedious tasks through achievements, rewards, and profile customization. A preliminary prototype evaluation was conducted with a small group of users to assess its usability and the impact of gamification on the GUI testing process. The results indicate high usability and positive reception of the gamification elements. However, due to the limited sample size of participants, further research is necessary to understand the plugin's effectiveness fully., Comment: 5 pages, 3 figures

Given the emergent reasoning abilities of large language models, information retrieval is becoming more complex. Rather than just retrieve a document, modern information retrieval systems advertise that they can synthesize an answer based on potentially many different documents, conflicting data sources, and using reasoning. We review recent literature and argue that the large language model has crucial flaws that prevent it from on its own ever constituting general intelligence, or answering general information synthesis requests. This review shows that the following are problems for large language models: hallucinations, complex reasoning, planning under uncertainty, and complex calculations. We outline how logical discrete graphical models can solve all of these problems, and outline a method of training a logical discrete model from unlabeled text., Comment: arXiv admin note: text overlap with arXiv:2402.06557