Search

The entanglement spectrum serves as a powerful tool for probing the structure and dynamics of quantum many-body systems, revealing key information about symmetry, topology, and excitations. While the entanglement entropy (EE) of ground states typically follows an area law, highly excited states obey a volume law, leading to a striking contrast in their scaling behavior. In this paper, we investigate the crossover between these two regimes, focusing on the role of quasi-particles (QPs) in mediating this transition. By analyzing the energy dependence of EE in various many-body systems, we explore how the presence of long-lived QPs influences the entanglement structure of excited states. We present numerical results for spinless fermions, a spin chain near a many-body localization transition, and the Sachdev-Ye-Kitaev (SYK) model, which lacks a conventional QP description. Our findings are complemented by a theoretical model based on Fermi liquid theory, providing insight into the interaction-dependent scaling of EE and its consistency with numerical simulations. We find that a hallmark of QPs is a linear dependence of the eigenstate EE on energy, which breaks down at high energy and in the limit of strong interaction. The slope of this linear dependence reflects the QP weight, which reduces with interaction strength., Comment: 8 pages, 6 figures

Vision-Language Models (VLMs) excel in diverse visual tasks but face challenges in document understanding, which requires fine-grained text processing. While typical visual tasks perform well with low-resolution inputs, reading-intensive applications demand high-resolution, resulting in significant computational overhead. Using OCR-extracted text in VLM prompts partially addresses this issue but underperforms compared to full-resolution counterpart, as it lacks the complete visual context needed for optimal performance. We introduce DocVLM, a method that integrates an OCR-based modality into VLMs to enhance document processing while preserving original weights. Our approach employs an OCR encoder to capture textual content and layout, compressing these into a compact set of learned queries incorporated into the VLM. Comprehensive evaluations across leading VLMs show that DocVLM significantly reduces reliance on high-resolution images for document understanding. In limited-token regimes (448$\times$448), DocVLM with 64 learned queries improves DocVQA results from 56.0% to 86.6% when integrated with InternVL2 and from 84.4% to 91.2% with Qwen2-VL. In LLaVA-OneVision, DocVLM achieves improved results while using 80% less image tokens. The reduced token usage allows processing multiple pages effectively, showing impressive zero-shot results on DUDE and state-of-the-art performance on MP-DocVQA, highlighting DocVLM's potential for applications requiring high-performance and efficiency.

Object detection is a significant field in autonomous driving. Popular sensors for this task include cameras and LiDAR sensors. LiDAR sensors offer several advantages, such as insensitivity to light changes, like in a dark setting and the ability to provide 3D information in the form of point clouds, which include the ranges of objects. However, 3D detection methods, such as PointPillars, typically require high-power hardware. Additionally, most common spinning LiDARs are sparse and may not achieve the desired quality of object detection in front of the car. In this paper, we present the feasibility of performing real-time 3D object detection of cars using 3D point clouds from a LiDAR sensor, processed and deployed on a low-power Hailo-8 AI accelerator. The LiDAR sensor used in this study is the InnovizOne sensor, which captures objects in higher quality compared to spinning LiDAR techniques, especially for distant objects. We successfully achieved real-time inference at a rate of approximately 5Hz with a high accuracy of 0.91% F1 score, with only -0.2% degradation compared to running the same model on an NVIDIA GeForce RTX 2080 Ti. This work demonstrates that effective real-time 3D object detection can be achieved on low-cost, low-power hardware, representing a significant step towards more accessible autonomous driving technologies. The source code and the pre-trained models are available at https://github.com/AIROTAU/ PointPillarsHailoInnoviz/tree/main

Superconductivity in doped SrTiO$_3$ has remained an enduring mystery for over 50 years. The material's status as a ``quantum" ferroelectric metal, characterized by a soft polar mode, suggests that quantum criticality could play a pivotal role in the emergence of its superconducting state. We show that the system is amenable to a strong coupling (Eliashberg) pairing analysis, with the dominant coupling to the soft mode being a ``dynamical'' Rashba coupling. We compute the expected $T_c$ for the entire phase diagram, all the way to the quantum critical point and beyond. We demonstrate that the linear coupling is sufficient to obtain a rough approximation of the experimentally measured phase diagram, but that nonlinear coupling terms are crucial in reproducing the finer features in the ordered phase. The primary role of nonlinear terms at the peak of the superconducting dome is to enhance the effective linear coupling induced by the broken order, shifting the dome's maximum into the ordered phase. Our theory quantitatively reproduces the three-dimensional experimental phase diagram in the space of carrier density, distance from the quantum critical point and temperature, and allows us to estimate microscopic parameters from the experimental data., Comment: 24 pages, 12 figures

Test-driven development (TDD) is the practice of writing tests first and coding later, and the proponents of TDD expound its numerous benefits. For instance, given an issue on a source code repository, tests can clarify the desired behavior among stake-holders before anyone writes code for the agreed-upon fix. Although there has been a lot of work on automated test generation for the practice "write code first, test later", there has been little such automation for TDD. Ideally, tests for TDD should be fail-to-pass (i.e., fail before the issue is resolved and pass after) and have good adequacy with respect to covering the code changed during issue resolution. This paper introduces TDD-Bench Verified, a high-quality benchmark suite of 449 issues mined from real-world GitHub code repositories. The benchmark's evaluation harness runs only relevant tests in isolation for simple yet accurate coverage measurements, and the benchmark's dataset is filtered both by human judges and by execution in the harness. This paper also presents Auto-TDD, an LLM-based solution that takes as input an issue description and a codebase (prior to issue resolution) and returns as output a test that can be used to validate the changes made for resolving the issue. Our evaluation shows that Auto-TDD yields a better fail-to-pass rate than the strongest prior work while also yielding high coverage adequacy. Overall, we hope that this work helps make developers more productive at resolving issues while simultaneously leading to more robust fixes.

Bacterial swarming is a complex phenomenon in which thousands of self-propelled rod-shaped cells move coherently on surfaces, providing an excellent example of active matter. However, bacterial swarming is different from most studied examples of active systems because single isolated cells do not move, while clusters do. The biophysical aspects underlying this behavior are unclear. In this work we explore the case of low local cell densities, where single cells become temporarily immobile. We show that immobility is related to local depletion of liquid. In addition, it is also associated with the state of the flagella. Specifically, the flagellar bundles at (temporarily) liquid depleted regions are completely spread-out. Our results suggest that dry models of self-propelled agents, which only consider steric alignments and neglect hydrodynamic effects, are oversimplified and are not sufficient to describe swarming bacteria.

This work is a sequel to our two 2023 publications [PLB 837 137669, NPA 1039 122725] where fitting 14 1$s_\Lambda$ and 1$p_\Lambda$ single-particle binding energies in hypernuclei across the periodic table led to a well-defined $\Lambda$-nucleus optical potential. The potential consists of a Pauli modified linear-density ($\Lambda N$) and a quadratic-density ($\Lambda NN$) terms. The present work reports on extending the above analysis to 21 $\Lambda$ single-particle data points input by including 1$d_\Lambda$ and 1$f_\Lambda$ states in medium-weight and heavy hypernuclei. The upgraded results for the $\Lambda N$ and $\Lambda NN$ potential depths at nuclear-matter density $\rho_0=0.17$~fm$^{-3}$, $D^{(2)}_\Lambda=-37.5\mp 0.7$~MeV and $D^{(3)}_\Lambda=+9.8\pm 1.2$~MeV together with the total depth $D_\Lambda=-27.7\pm 0.5$~MeV, agree within errors with the earlier results. The $\Lambda$ hypernuclear overbinding associated with the $\Lambda N$-induced potential depth $D^{(2)}_\Lambda$ agrees quantitatively with a recent combined analysis of low-energy $\Lambda p$ scattering data and correlation functions [PLB 850 (2024) 138550]. These results, particularly the size of the repulsive $D^{(3)}_\Lambda$, provide an essential input towards resolving the 'hyperon puzzle' in the core of neutron stars. We also show that a key property of our $\Lambda NN$-induced potential term, i.e. a need to suppress the quadratic-density $\Lambda NN$ term involving an excess neutron and a $N=Z$ core nucleon, can be tested in the forthcoming JLab E12-15-008 experiment., Comment: 10 pages, 3 figures, 1 table. Submitted to Proceedings of International Conference on Exotic Atoms and Related Topics and Conference on Low Energy Antiprotons (EXA-LEAP2024)

ABC-stacked rhombohedral graphene multilayers exhibit a wide variety of electronic ground states characterized by broken isospin symmetry and superconductivity. Recently, indirect evidence of inter-valley coherent (IVC) order has been reported in rhombohedral trilayer graphene (RTG), with possible implications for the origin of superconductivity. Here, we report the direct visualization of IVC order in RTG using scanning tunneling microscopy and spectroscopy. Tuning the chemical potential through the Van Hove singularity near the edge of the valence band, we observe a cascade of phase transitions associated with the formation of half- and quarter-metal states. IVC phases, distinguished by an enlarged real space unit cell, are directly imaged near both the high- and low-density boundaries of the half-metal phase. At high hole density, we precisely reconstruct the IVC band structure through quasiparticle interference. Intriguingly, the charge density modulations reveal a C3-symmetric incommensurate IVC order that agrees with the recent prediction of an IVC-crystal phase. Our findings demonstrate that IVC phases are a widespread symmetry-broken ground state within graphene systems., Comment: 22 pages, 4 figures

Vision-Language (VL) models have garnered considerable research interest; however, they still face challenges in effectively handling text within images. To address this limitation, researchers have developed two approaches. The first method involves utilizing external Optical Character Recognition (OCR) tools to extract textual information from images, which is then prepended to other textual inputs. The second strategy focuses on employing extremely high-resolution images to improve text recognition capabilities. In this paper, we focus on enhancing the first strategy by introducing a novel method, named TAP-VL, which treats OCR information as a distinct modality and seamlessly integrates it into any VL model. TAP-VL employs a lightweight transformer-based OCR module to receive OCR with layout information, compressing it into a short fixed-length sequence for input into the LLM. Initially, we conduct model-agnostic pretraining of the OCR module on unlabeled documents, followed by its integration into any VL architecture through brief fine-tuning. Extensive experiments demonstrate consistent performance improvements when applying TAP-VL to top-performing VL models, across scene-text and document-based VL benchmarks.

Readers can have different goals with respect to the text they are reading. Can these goals be decoded from the pattern of their eye movements over the text? In this work, we examine for the first time whether it is possible to decode two types of reading goals that are common in daily life: information seeking and ordinary reading. Using large scale eye-tracking data, we apply to this task a wide range of state-of-the-art models for eye movements and text that cover different architectural and data representation strategies, and further introduce a new model ensemble. We systematically evaluate these models at three levels of generalization: new textual item, new participant, and the combination of both. We find that eye movements contain highly valuable signals for this task. We further perform an error analysis which builds on prior empirical findings on differences between ordinary reading and information seeking and leverages rich textual annotations. This analysis reveals key properties of textual items and participant eye movements that contribute to the difficulty of the task.

Shared control problems involve a robot learning to collaborate with a human. When learning a shared control policy, short communication between the agents can often significantly reduce running times and improve the system's accuracy. We extend the shared control problem to include the ability to directly query a cooperating agent. We consider two types of potential responses to a query, namely oracles: one that can provide the learner with the best action they should take, even when that action might be myopically wrong, and one with a bounded knowledge limited to its part of the system. Given this additional information channel, this work further presents three heuristics for choosing when to query: reinforcement learning-based, utility-based, and entropy-based. These heuristics aim to reduce a system's overall learning cost. Empirical results on two environments show the benefits of querying to learn a better control policy and the tradeoffs between the proposed heuristics.

Black box attacks, where adversaries have limited knowledge of the target model, pose a significant threat to machine learning systems. Adversarial examples generated with a substitute model often suffer from limited transferability to the target model. While recent work explores ranking perturbations for improved success rates, these methods see only modest gains. We propose a novel strategy called PEAS that can boost the transferability of existing black box attacks. PEAS leverages the insight that samples which are perceptually equivalent exhibit significant variability in their adversarial transferability. Our approach first generates a set of images from an initial sample via subtle augmentations. We then evaluate the transferability of adversarial perturbations on these images using a set of substitute models. Finally, the most transferable adversarial example is selected and used for the attack. Our experiments show that PEAS can double the performance of existing attacks, achieving a 2.5x improvement in attack success rates on average over current ranking methods. We thoroughly evaluate PEAS on ImageNet and CIFAR-10, analyze hyperparameter impacts, and provide an ablation study to isolate each component's importance.

Can human reading comprehension be assessed from eye movements in reading? In this work, we address this longstanding question using large-scale eyetracking data over textual materials that are geared towards behavioral analyses of reading comprehension. We focus on a fine-grained and largely unaddressed task of predicting reading comprehension from eye movements at the level of a single question over a passage. We tackle this task using three new multimodal language models, as well as a battery of prior models from the literature. We evaluate the models' ability to generalize to new textual items, new participants, and the combination of both, in two different reading regimes, ordinary reading and information seeking. The evaluations suggest that although the task is highly challenging, eye movements contain useful signals for fine-grained prediction of reading comprehension. Code and data will be made publicly available., Comment: Accepted to EMNLP

Materials hosting flat bands at the vicinity of the Fermi level promote exotic symmetry broken states. Common to many of these are van Hove singularities at saddle points of the dispersion or even higher-order van Hove singularities where the dispersion is flattened further. The band structure of kagome metals hosts both a flat band and two regular saddle points flanking a Dirac node. We investigate the kagome ferromagnetic metal Co$_3$Sn$_2$S$_2$ using scanning tunneling spectroscopy. We identify a new mechanism by which a triangular distortion on its kagome Co$_3$Sn surface termination considerably flattens the saddle point dispersion, and induces an isolated higher-order van Hove singularity (HOvHS) with algebraically divergent density of states pinned to the Fermi energy. The distortion-induced HOvHS precipitates a Pomeranchuk instability of the Fermi surface, resulting in the formation of a series of nematic electronic states. We visualize the nematic order across an energy shell of about 100 meV in both real-, reciprocal-, and momentum-spaces, as a cascade of wavefunction distributions which spontaneously break the remaining rotational symmetry of the underlying distorted kagome lattice, without generating any additional translational symmetry breaking. It signifies the spontaneous removal of a subset of saddle points from the Fermi energy to lower energies. By tracking the electronic wavefunction structure across the deformed Fermi surface we further identify a charge pumping-like evolution of the wavefunction center of mass. The mechanism we find for the generation of higher-order saddle points under a kagome distortion may be common to other kagome materials, and potentially other lattice structures, suggesting a generic new avenue for inducing unconventional electronic instabilities towards exotic states of matter.

Following significant progress in the visualization and characterization of hybrid superconducting-semiconducting systems, greatly propelled by reports of Majorana zero modes in nanowire devices, considerable attention has been devoted to investigating the electronic structure at the buried superconducting-semiconducting interface and the nature of the induced superconducting correlations. The properties of that interface and the structure of the electronic wave functions that occupy it determine the functionality and the topological nature of the induced superconducting state. Here, we introduce a novel hybrid platform for proximity-inducing superconductivity in InAs$_{0.6}$Sb$_{0.4}$ nanowires, leveraging a unique architecture and material combination. By dispersing these nanowires over a superconducting Indium film we exploit Indium's high critical temperature of 3.7~K and the anticipated high spin-orbit and Zeeman couplings of InAs$_{0.6}$Sb$_{0.4}$. This design preserves the pristine top facet of the nanowires, making it highly compatible with scanning tunneling spectroscopy. Using this architecture we demonstrate that the mechanical contact supports Cooper-pair transparency as high as 90\%, comparable with epitaxial interfaces. The anisotropic angular response to an applied magnetic field shows the quasi-two-dimensional nature of the parent superconductivity in the Indium film and the quasi-one-dimensional nature of the induced superconductivity in the nanowires. Our platform offers robust and advantageous foundations for studying the emergence of topological superconductivity and the interplay of superconductivity and magnetism using atomic-scale spectroscopic tools., Comment: 30 pages, 20 figures

An essential step in building a quantum computer system is combining two qubits. Previous theoretical analyses of this element have been based primarily on the short schematic matrix approaches. In the present work we extend the investigation to the combination of two qubits by implementing the Dirac picture method as a theoretical framework. We conduct simulations, using parameters of grAl qubits, to demonstrate the feasibility of driving two qubits into an even-balanced quantum state. The research highlights the effectiveness of the Dirac picture method in achieving the desired state, enabling improved control of entangled states and advanced quantum computing research. The simulation results and the theoretical calculations agree within of 0.2%, and the two examples at section IV, show a good agreement to the level of 6% error from the even balanced state, Comment: 7 pages, 9 figures

Training a fine-grained image recognition model with limited data presents a significant challenge, as the subtle differences between categories may not be easily discernible amidst distracting noise patterns. One commonly employed strategy is to leverage pretrained neural networks, which can generate effective feature representations for constructing an image classification model with a restricted dataset. However, these pretrained neural networks are typically trained for different tasks than the fine-grained visual recognition (FGVR) task at hand, which can lead to the extraction of less relevant features. Moreover, in the context of building FGVR models with limited data, these irrelevant features can dominate the training process, overshadowing more useful, generalizable discriminative features. Our research has identified a surprisingly simple solution to this challenge: we introduce a regularization technique to ensure that the magnitudes of the extracted features are evenly distributed. This regularization is achieved by maximizing the uniformity of feature magnitude distribution, measured through the entropy of the normalized features. The motivation behind this regularization is to remove bias in feature magnitudes from pretrained models, where some features may be more prominent and, consequently, more likely to be used for classification. Additionally, we have developed a dynamic weighting mechanism to adjust the strength of this regularization throughout the learning process. Despite its apparent simplicity, our approach has demonstrated significant performance improvements across various fine-grained visual recognition datasets.

Charged particle reconstruction is one the most computationally heavy components of the full event reconstruction of Large Hadron Collider (LHC) experiments. Looking to the future, projections for the High Luminosity LHC (HL-LHC) indicate a superlinear growth for required computing resources for single-threaded CPU algorithms that surpass the computing resources that are expected to be available. The combination of these facts creates the need for efficient and computationally performant pattern recognition algorithms that will be able to run in parallel and possibly on other hardware, such as GPUs, given that these become more and more available in LHC experiments and high-performance computing centres. Line Segment Tracking (LST) is a novel such algorithm which has been developed to be fully parallelizable and hardware agnostic. The latter is achieved through the usage of the Alpaka library. The LST algorithm has been tested with the CMS central software as an external package and has been used in the context of the CMS HL-LHC High Level Trigger (HLT). When employing LST for pattern recognition in the HLT tracking, the physics and timing performances are shown to improve with respect to the ones utilizing the current pattern recognition algorithms. The latest results on the usage of the LST algorithm within the CMS HL-LHC HLT are presented, along with prospects for further improvements of the algorithm and its CMS central software integration.

We present an efficient method for adapting a monolingual Large Language Model (LLM) to another language, addressing challenges of catastrophic forgetting and tokenizer limitations. We focus this study on adapting Llama 2 to Arabic. Our two-stage approach begins with expanding the vocabulary and training only the embeddings matrix, followed by full model continual pre-training on a bilingual corpus. By continually pre-training on a mix of Arabic and English corpora, the model retains its proficiency in English while acquiring capabilities in Arabic. Our approach results in significant improvements in Arabic and slight enhancements in English, demonstrating cost-effective cross-lingual transfer. We perform ablations on embedding initialization techniques, data mix ratios, and learning rates and release a detailed training recipe. To demonstrate generalizability of this approach we also adapted Llama 3 8B to Arabic and Llama 2 13B to Hindi.

Deep neural networks (DNNs) are widely used in real-world applications, yet they remain vulnerable to errors and adversarial attacks. Formal verification offers a systematic approach to identify and mitigate these vulnerabilities, enhancing model robustness and reliability. While most existing verification methods focus on image classification models, this work extends formal verification to the more complex domain of emph{object detection} models. We propose a formulation for verifying the robustness of such models and demonstrate how state-of-the-art verification tools, originally developed for classification, can be adapted for this purpose. Our experiments, conducted on various datasets and networks, highlight the ability of formal verification to uncover vulnerabilities in object detection models, underscoring the need to extend verification efforts to this domain. This work lays the foundation for further research into formal verification across a broader range of computer vision applications.

It is now established that glasses feature low-frequency, nonphononic excitations, in addition to phonons that follow Debye's vibrational density of state (VDoS). Extensive computer studies demonstrated that these nonphononic, glassy excitations follow a universal non-Debye VDoS ${\cal D}_{\rm G}(\omega)\!\sim\!\omega^4$, at low frequencies $\omega$. Yet, due to intrinsic difficulties in disentangling ${\cal D}_{\rm G}(\omega)$ from the total VDoS ${\cal D}(\omega)$, which is experimentally accessible through various scattering techniques, the $\omega^4$ tail of ${\cal D}_{\rm G}(\omega)$ lacked direct experimental support. We develop a procedure to extract ${\cal D}_{\rm G}(\omega)$ from the measured ${\cal D}(\omega)$, based on recent advances in understanding low-frequency excitations in glasses, and apply it to available datasets for diverse glasses. The resulting analysis shows that the $\omega^4$ tail of the nonphononic vibrational spectra of glasses is nontrivially consistent with a broad range of experimental observations. It also further supports that ${\cal D}_{\rm G}(\omega)$ makes an additive contribution to ${\cal D}(\omega)$., Comment: v2: parameter-free support to the main findings was added (new Fig. 4); see related work at: https://doi.org/10.1063/5.0216351

We study to what extent the unique observation of $\Lambda\Lambda$ hypernuclei by their weak decay into known $\Lambda$ hypernuclei, with lifetimes of order 10$^{-10}$ s, rules out the existence of a deeply bound doubly-strange (${\cal S}$=$-$2) $H$ dibaryon. Treating ${_{\Lambda\Lambda}^{~~6}}{\rm He}$ (the Nagara emulsion event) in a realistic $\Lambda-\Lambda-{^4}$He three-body model, we find that the ${_{\Lambda\Lambda}^{~~6}}{\rm He}\to H + {^4{\rm He}}$ strong-interaction lifetime increases beyond 10$^{-10}$ s for $m_H < m_{\Lambda}+m_n$, about 176 MeV below the $\Lambda\Lambda$ threshold, so that such a deeply bound $H$ is not in conflict with hypernuclear data. Constrained by $\Lambda$ hypernuclear $\Delta{\cal S}$=1 nonmesonic weak-interaction decay rates, we follow EFT methods to evaluate the $\Delta{\cal S}$=2 $H\to nn$ weak-decay lifetime of $H$ in the mass range $2m_n \lesssim m_H < m_{\Lambda}+m_n$. The resulting $H$ lifetime is of order 10$^5$ s, many orders of magnitude shorter than required to qualify for a dark-matter candidate., Comment: v1: 17 pages, 2 figures, 3 tables; v2: slightly edited, Table 3 corrected, submitted for publication; v3: slightly edited in response to referee's comments, 19 pages, 2 figures, 3 tables, PLB published version

In addition to Goldstone phonons that generically emerge in the low-frequency vibrational spectrum of any solid, crystalline or glassy, structural glasses also feature other low-frequency vibrational modes. The nature and statistical properties of these modes -- often termed `excess modes' -- have been the subject of decades-long investigation. Studying them, even using well-controlled computer glasses, has proven challenging due to strong spatial hybridization effects between phononic and nonphononic excitations, which hinder quantitative analyses of the nonphononic contribution ${\cal D}_{\rm G}(\omega)$ to the total spectrum ${\cal D}(\omega)$, per frequency $\omega$. Here, using recent advances indicating that ${\cal D}_{\rm G}(\omega)\!=\!{\cal D}(\omega)-{\cal D}_{\rm D}(\omega)$, where ${\cal D}_{\rm D}(\omega)$ is Debye's spectrum of phonons, we present a simple and straightforward scheme to enumerate nonphononic modes in computer glasses. Our analysis establishes that nonphononic modes in computer glasses indeed make an additive contribution to the total spectrum, including in the presence of strong hybridizations. Moreover, it cleanly reveals the universal ${\cal D}_{\rm G}(\omega)\!\sim\!\omega^4$ tail of the nonphononic spectrum, and opens the way for related analyses of experimental spectra of glasses., Comment: 7 pages, 5 figures. V2: a reference added to a closely related analysis of experimental data, see arXiv:2404.16996

The electron-electron and electron-phonon coupling in complex materials can be more complicated than simple density-density interactions, involving intertwined dynamics of spin, charge, and spatial symmetries. This motivates studying universal models with complex interactions, and studying whether in this case BCS-type singlet pairing is still the ``natural'' fate of the system. To this end, we construct a Yukawa-SYK model with nonlocal couplings in both spin and charge channels. Furthermore, we provide for time-reversal-symmetry breaking dynamics by averaging over the Gaussian Unitary ensemble rather than the Orthogonal ensemble. We find that the ground state of the system can be an orbitally nonlocal superconducting state arising from incoherent fermions with no BCS-like analog. The superconductivity has an equal tendency to triplet and singlet pairing states separated by a non-Fermi liquid phase. We further study the fate of the system within the superconducting phase and find that the expected ground state, away from the critical point, is a mixed singlet/triplet state. Finally, we find that while at $T_c$ the triplet and singlet transitions are dual to one another, below $T_c$ the duality is broken, with the triplet state more susceptible to orbital fluctuations just by virtue of its symmetry. Our results indicate that such fluctuation-induced mixed states may be an inherent feature of strongly correlated materials., Comment: 4 pages, 3 figures

We present an improvement of the phase distance correlation (PDC) periodogram to account for uncertainties in the time-series data. The PDC periodogram introduced in our previous papers is based on the statistical concept of distance correlation. By viewing each measurement and its accompanying error estimate as a probability distribution, we are able to use the concept of energy distance to design a distance function (metric) between measurement-uncertainty pairs. We used this metric as the basis for the PDC periodogram,instead of the simple absolute difference. We demonstrate the periodogram's performance using both simulated and real-life data. This adaptation makes the PDC periodogram much more useful, demonstrating it can be helpful in the exploration of large time-resolved astronomical databases, ranging from Gaia radial velocity and photometry data releases to those of smaller surveys, such as APOGEE and LAMOST. We have made a public GitHub repository available, with a Python implementation of the new tools available to the community., Comment: 6 pages, 6 figures. A&A accepted

We introduce a notion of distributions on $\mathbb{R}^n$, called distributions of C$^{{\mathrm{exp}}}$-class, based on wavelet transforms of distributions and the theory from [6] about C$^{{\mathrm{exp}}}$-class functions. We prove that the framework of C$^{{\mathrm{exp}}}$-class distributions is closed under natural operations, like push-forward, pull-back, derivation and antiderivation, and, in the tempered case, Fourier transforms. Our main result is the (real analytic) holonomicity of all distributions of C$^{{\mathrm{exp}}}$-class.

Novel-view synthesis through diffusion models has demonstrated remarkable potential for generating diverse and high-quality images. Yet, the independent process of image generation in these prevailing methods leads to challenges in maintaining multiple-view consistency. To address this, we introduce ViewFusion, a novel, training-free algorithm that can be seamlessly integrated into existing pre-trained diffusion models. Our approach adopts an auto-regressive method that implicitly leverages previously generated views as context for the next view generation, ensuring robust multi-view consistency during the novel-view generation process. Through a diffusion process that fuses known-view information via interpolated denoising, our framework successfully extends single-view conditioned models to work in multiple-view conditional settings without any additional fine-tuning. Extensive experimental results demonstrate the effectiveness of ViewFusion in generating consistent and detailed novel views., Comment: CVPR2024,homepage:https://wi-sc.github.io/ViewFusion.github.io/

Vision-Language (VL) models have gained significant research focus, enabling remarkable advances in multimodal reasoning. These architectures typically comprise a vision encoder, a Large Language Model (LLM), and a projection module that aligns visual features with the LLM's representation space. Despite their success, a critical limitation persists: the vision encoding process remains decoupled from user queries, often in the form of image-related questions. Consequently, the resulting visual features may not be optimally attuned to the query-specific elements of the image. To address this, we introduce QA-ViT, a Question Aware Vision Transformer approach for multimodal reasoning, which embeds question awareness directly within the vision encoder. This integration results in dynamic visual features focusing on relevant image aspects to the posed question. QA-ViT is model-agnostic and can be incorporated efficiently into any VL architecture. Extensive experiments demonstrate the effectiveness of applying our method to various multimodal architectures, leading to consistent improvement across diverse tasks and showcasing its potential for enhancing visual and scene-text understanding.

Materials typically fail under complex stress states, essentially involving dilatational (volumetric) components that eventually lead to material decohesion/separation. It is therefore important to understand dilatational irreversible deformation -- i.e., dilatational plasticity -- en route to failure. In the context of glasses, much focus has been given to shear (volume-preserving) plasticity, both in terms of the stress states considered and the corresponding material response. Here, using a recently-developed methodology and extensive computer simulations, we shed basic light on the elementary processes mediating dilatational plasticity in glasses. We show that plastic instabilities, corresponding to singularities of the glass Hessian, generically feature both dilatational and shear irreversible strain components. The relative magnitude and statistics of the strain components depend both on the symmetry of the driving stress (e.g., shear vs.~hydrostatic tension) and on the cohesive (attractive) part of the interatomic interaction. We further show that the tensorial shear component of the plastic strain is generally non-planar and also extract the characteristic volume of plastic instabilities. Elucidating the fundamental properties of the elementary micro-mechanical building blocks of plasticity in glasses sets the stage for addressing larger-scale, collective phenomena in dilatational plasticity such as topological changes in the form of cavitation and ductile-to-brittle transitions. As a first step in this direction, we show that the elastic moduli markedly soften during dilatational plastic deformation approaching cavitation., Comment: Slightly revised presentation, no change in content, fixed bibliographical list

Neural radiance fields (NeRFs) have exhibited potential in synthesizing high-fidelity views of 3D scenes but the standard training paradigm of NeRF presupposes an equal importance for each image in the training set. This assumption poses a significant challenge for rendering specific views presenting intricate geometries, thereby resulting in suboptimal performance. In this paper, we take a closer look at the implications of the current training paradigm and redesign this for more superior rendering quality by NeRFs. Dividing input views into multiple groups based on their visual similarities and training individual models on each of these groups enables each model to specialize on specific regions without sacrificing speed or efficiency. Subsequently, the knowledge of these specialized models is aggregated into a single entity via a teacher-student distillation paradigm, enabling spatial efficiency for online render-ing. Empirically, we evaluate our novel training framework on two publicly available datasets, namely NeRF synthetic and Tanks&Temples. Our evaluation demonstrates that our DaC training pipeline enhances the rendering quality of a state-of-the-art baseline model while exhibiting convergence to a superior minimum.

It has been shown that the spontaneous emission rate of photons by free electrons, unlike stimulated emission, is independent of the shape or modulation of the quantum electron wavefunction (QEW). Nevertheless, here we show that the quantum state of the emitted photons is non-classical and does depend on the QEW shape. This non-classicality originates from the shape dependent off-diagonal terms of the photon density matrix. This is manifested in the Wigner distribution function and would be observable experimentally through Homodyne detection techniques as a squeezing effect. Considering a scheme of electrons interaction with a single microcavity mode, we present a QED formulation of spontaneous emission by multiple modulated QEWs through a build-up process. Our findings indicate that in the case of a density modulated QEWs beam, the phase of the off-diagonal terms of the photon state emitted by the modulated QEWs is the harbinger of bunched beam superradiance, where the spontaneous emission is proportional to N_e^2. This observation offers a potential for enhancement of other quantum electron interactions with quantum systems by a modulated QEWs beam carrying coherence and quantum properties of the modulation., Comment: 52 pages, 6 figures,Suppl. Mater. 25 pages, Suppl. Mater. 5 figures, 1 table

Topological defects, which are singular points in a director field, play a major role in shaping active systems. Here, we experimentally study topological defects and the flow patterns around them, that are formed during the highly rapid dynamics of swarming bacteria. The results are compared to the predictions of two-dimensional active nematics. We show that, even though some of the assumptions underlying the theory do not hold, the swarm dynamics is in agreement with two-dimensional nematic theory. In particular, we look into the multi-layered structure of the swarm, which is an important feature of real, natural colonies, and find a strong coupling between layers. Our results suggest that the defect-charge density is hyperuniform, i.e., that long range density-fluctuations are suppressed., Comment: 9 pages, 7 figures

We investigate a two-period Bayesian persuasion game, where the receiver faces a decision, akin to a one-armed bandit problem: to undertake an action, gaining noisy information and a corresponding positive or negative payoff, or to refrain. The sender's objective is to dissuade the receiver from taking action by furnishing information about the payoff. Our findings describe the optimal strategy for the amount and timing of information disclosure. In scenarios where the sender possesses knowledge of the receiver's first-period action or observes a noisy public signal correlated with it, the optimal strategy entails revealing information in the second period. If this alone proves to be insufficient to dissuade the receiver from acting, supplementary information is provided in the first period. In scenarios where information must be provided without conditioning on the receiver's first-period action, the optimal strategy entails revealing information exclusively in the first period.

The increasing use of transformer-based large language models brings forward the challenge of processing long sequences. In document visual question answering (DocVQA), leading methods focus on the single-page setting, while documents can span hundreds of pages. We present GRAM, a method that seamlessly extends pre-trained single-page models to the multi-page setting, without requiring computationally-heavy pretraining. To do so, we leverage a single-page encoder for local page-level understanding, and enhance it with document-level designated layers and learnable tokens, facilitating the flow of information across pages for global reasoning. To enforce our model to utilize the newly introduced document tokens, we propose a tailored bias adaptation method. For additional computational savings during decoding, we introduce an optional compression stage using our compression-transformer (C-Former),reducing the encoded sequence length, thereby allowing a tradeoff between quality and latency. Extensive experiments showcase GRAM's state-of-the-art performance on the benchmarks for multi-page DocVQA, demonstrating the effectiveness of our approach.

In this paper, we focus on the prediction phase of a random forest and study the problem of representing a bag of decision trees using a smaller bag of decision trees, where we only consider binary decision problems on the binary domain and simple decision trees in which an internal node is limited to querying the Boolean value of a single variable. As a main result, we show that the majority function of $n$ variables can be represented by a bag of $T$ ($< n$) decision trees each with polynomial size if $n-T$ is a constant, where $n$ and $T$ must be odd (in order to avoid the tie break). We also show that a bag of $n$ decision trees can be represented by a bag of $T$ decision trees each with polynomial size if $n-T$ is a constant and a small classification error is allowed. A related result on the $k$-out-of-$n$ functions is presented too.

Acoustic-directed assembly is a modular and flexible bottom-up technique with the potential to pattern a wide range of materials. Standing acoustic waves have been previously employed for patterning preformed metal particles, however, direct patterning of metallic structures from precursors remains unexplored. Here, we investigate utilization of standing waves to exert control over chemical reaction products, while also exploring their potential in the formation of multi-layered and composite micro-structures. Periodic metallic micro-structures were formed in a single step, simplifying microstructure fabrication. Concentric structures were obtained by introducing a metal precursor salt and a reducing agent into a cylindrical piezoelectric resonator that also served as a reservoir. In addition, we introduce an innovative approach to directly fabricate metallic multi-layer and composite structures by reducing different metal ions or adding nanoparticles during the reduction step. Fewer steps are needed, compared with other methods, and there is no need to stabilize the nanoparticles or to ensure chemical affinity between the metallic matrix and inorganic nanoparticles. This innovative approach is promising for production of complex microstructures with enhanced functionality and controlled properties.