Search

We revisit particle creation in strong fields, and backreaction on those fields, from an amplitudes perspective. We describe the strong field by an initial coherent state of photons which we explicitly evolve in time, thus going beyond the background field approximation, and then consider observables which quantify the effects of backreaction. We present expressions for the waveform, vacuum persistence probability, and number of produced photons at next-to-leading order, all of which are impacted by backreaction, along with the number and statistics of produced pairs. We find that converting between in-out (amplitude) and in-in (expectation value) expressions requires explicit resummation of an infinite number of disconnected loop diagrams., Comment: 18 pages

Large language models (LLMs) often fail to synthesize information from their context to generate an accurate response. This renders them unreliable in knowledge intensive settings where reliability of the output is key. A critical component for reliable LLMs is the integration of a robust fact-checking system that can detect hallucinations across various formats. While several open-access fact-checking models are available, their functionality is often limited to specific tasks, such as grounded question-answering or entailment verification, and they perform less effectively in conversational settings. On the other hand, closed-access models like GPT-4 and Claude offer greater flexibility across different contexts, including grounded dialogue verification, but are hindered by high costs and latency. In this work, we introduce VERITAS, a family of hallucination detection models designed to operate flexibly across diverse contexts while minimizing latency and costs. VERITAS achieves state-of-the-art results considering average performance on all major hallucination detection benchmarks, with $10\%$ increase in average performance when compared to similar-sized models and get close to the performance of GPT4 turbo with LLM-as-a-judge setting.

Experience goods such as sporting and artistic events, songs, videos, news stories, podcasts, and television series, are often packaged and consumed in bundles. Many such bundles are ordered in the sense that the individual items are consumed sequentially, one at a time. We examine if an individual's decision to consume the next item in an ordered bundle can be predicted based on his/her consumption pattern for the preceding items. We evaluate several predictive models, including two custom Transformers using decoder-only and encoder-decoder architectures, fine-tuned GPT-3, a custom LSTM model, a reinforcement learning model, two Markov models, and a zero-order model. Using data from Spotify, we find that the custom Transformer with a decoder-only architecture provides the most accurate predictions, both for individual choices and aggregate demand. This model captures a general form of state dependence. Analysis of Transformer attention weights suggests that the consumption of the next item in a bundle is based on approximately equal weighting of all preceding choices. Our results indicate that the Transformer can assist in queuing the next item that an individual is likely to consume from an ordered bundle, predicting the demand for individual items, and personalizing promotions to increase demand.

In this letter, we investigate how field redefinition influences the spectrum of linearized perturbations in relativistic fluid dynamics. We show that the hydrodynamic modes do not get affected under local field redefinition, whereas the non-hydrodynamic modes do. These non-hydrodynamic modes can be removed through a suitable all-order field redefinition. This process leads to a new frame containing only hydrodynamic modes, which we refer to as the hydro frame. Additionally, we demonstrate that the resulting stress-energy tensor may constitute an infinite series in momentum space, with the radius of convergence associated with the removed non-hydrodynamic mode, highlighting its role in the hydrodynamic expansion's validity., Comment: 8 pages; Comments are welcome; Updated discussion and references

Deep neural networks trained on biased data often inadvertently learn unintended inference rules, particularly when labels are strongly correlated with biased features. Existing bias mitigation methods typically involve either a) predefining bias types and enforcing them as prior knowledge or b) reweighting training samples to emphasize bias-conflicting samples over bias-aligned samples. However, both strategies address bias indirectly in the feature or sample space, with no control over learned weights, making it difficult to control the bias propagation across different layers. Based on this observation, we introduce a novel approach to address bias directly in the model's parameter space, preventing its propagation across layers. Our method involves training two models: a bias model for biased features and a debias model for unbiased details, guided by the bias model. We enforce dissimilarity in the debias model's later layers and similarity in its initial layers with the bias model, ensuring it learns unbiased low-level features without adopting biased high-level abstractions. By incorporating this explicit constraint during training, our approach shows enhanced classification accuracy and debiasing effectiveness across various synthetic and real-world datasets of different sizes. Moreover, the proposed method demonstrates robustness across different bias types and percentages of biased samples in the training data. The code is available at: https://visdomlab.github.io/CosFairNet/

Network slicing allows Telecom Operators (TOs) to support service provisioning with diverse Service Level Agreements (SLAs). The combination of network slicing and Open Radio Access Network (RAN) enables TOs to provide more customized network services and higher commercial benefits. However, in the current Open RAN community, an open-source end-to-end slicing solution for 5G is still missing. To bridge this gap, we developed ORANSlice, an open-source network slicing-enabled Open RAN system integrated with popular open-source RAN frameworks. ORANSlice features programmable, 3GPP-compliant RAN slicing and scheduling functionalities. It supports RAN slicing control and optimization via xApps on the near-real-time RAN Intelligent Controller (RIC) thanks to an extension of the E2 interface between RIC and RAN, and service models for slicing. We deploy and test ORANSlice on different O-RAN testbeds and demonstrate its capabilities on different use cases, including slice prioritization and minimum radio resource guarantee.

One of the most challenging problems in the computational study of localization in quantum manybody systems is to capture the effects of rare events, which requires sampling over exponentially many disorder realizations. We implement an efficient procedure on a quantum processor, leveraging quantum parallelism, to efficiently sample over all disorder realizations. We observe localization without disorder in quantum many-body dynamics in one and two dimensions: perturbations do not diffuse even though both the generator of evolution and the initial states are fully translationally invariant. The disorder strength as well as its density can be readily tuned using the initial state. Furthermore, we demonstrate the versatility of our platform by measuring Renyi entropies. Our method could also be extended to higher moments of the physical observables and disorder learning.

While photons and gravitons do not interact significantly, photons can be converted to gravitons in a background magnetic field -- a phenomenon known as the Gertsenshtein effect. In this paper, we investigate whether chiral electromagnetic (EM) waves can be converted to chiral gravitational waves (GW) in the presence of primordial magnetic fields during the radiation-dominated epoch of the early universe. We consider two situations wherein chirality is either present in the propagating EM waves or it exists in the background magnetic field. Our analysis shows that while the conversion probability increases with stronger magnetic fields, it remains insensitive to the chiral nature of the background magnetic field. Consequently, the net chirality parameter is independent of the chirality of the background field in both cases. Finally, we demonstrate that the present-day energy density of the produced chiral GWs peaks at a frequency of $\sim 100$ GHz, and the corresponding characteristic strain will be highly sensitive to current and future missions designed to detect high-frequency GWs., Comment: v2: 28 pages, 4 figures, updated Fig.3

LLM-as-a-judge models have been used for evaluating both human and AI generated content, specifically by providing scores and rationales. Rationales, in addition to increasing transparency, help models learn to calibrate its judgments. Enhancing a model's rationale can therefore improve its calibration abilities and ultimately the ability to score content. We introduce Self-Rationalization, an iterative process of improving the rationales for the judge models, which consequently improves the score for fine-grained customizable scoring criteria (i.e., likert-scale scoring with arbitrary evaluation criteria). Self-rationalization works by having the model generate multiple judgments with rationales for the same input, curating a preference pair dataset from its own judgements, and iteratively fine-tuning the judge via DPO. Intuitively, this approach allows the judge model to self-improve by learning from its own rationales, leading to better alignment and evaluation accuracy. After just two iterations -- while only relying on examples in the training set -- human evaluation shows that our judge model learns to produce higher quality rationales, with a win rate of $62\%$ on average compared to models just trained via SFT on rationale . This judge model also achieves high scoring accuracy on BigGen Bench and Reward Bench, outperforming even bigger sized models trained using SFT with rationale, self-consistency or best-of-$N$ sampling by $3\%$ to $9\%$.

Implicit neural representations (INR) have been recently adopted in various applications ranging from computer vision tasks to physics simulations by solving partial differential equations. Among existing INR-based works, multi-layer perceptrons with sinusoidal activation functions find widespread applications and are also frequently treated as a baseline for the development of better activation functions for INR applications. Recent investigations claim that the use of sinusoidal activation functions could be sub-optimal due to their limited supported frequency set as well as their tendency to generate over-smoothed solutions. We provide a simple solution to mitigate such an issue by changing the activation function at the first layer from $\sin(x)$ to $\sin(\sinh(2x))$. We demonstrate H-SIREN in various computer vision and fluid flow problems, where it surpasses the performance of several state-of-the-art INRs.

In the context of fifth-generation new radio (5G NR) technology, it is not possible to directly obtain an absolute uplink (UL) channel impulse response (CIR) at the base station (gNB) from a user equipment (UE). The UL CIR obtained through the sounding reference signal (SRS) is always time-shifted by the timing advance (TA) applied at the UE. The TA is crucial for maintaining UL synchronization, and transmitting SRS without applying the TA will result in interference. In this work, we propose a new method to obtain absolute UL CIR from a UE and then use it to estimate the round trip time (RTT) at the gNB. This method requires enhancing the current 5G protocol stack with a new Zadoff-Chu (ZC) based wideband uplink reference signal (URS). Capitalizing on the cyclic shift property of the URS sequence, we can obtain the RTT with a significant reduction in overhead and latency compared to existing schemes. The proposed method is experimentally validated using a real-world testbed based on OpenAirInterface (OAI).

Solid-state electrolytes have the potential to stabilize lithium metal anodes, which hold the promise to increase the energy density of lithium-ion batteries. However, lithium metal dendrites that occur locally at the solid-solid interface plague the solid-state lithium metal cells during charging, limiting the cycle life, and rendering high safety risks in practical applications. While multiple explanations have been proposed, understandings of the dynamics preceding and causing the metal penetration in solid electrolytes are still not conclusive. Here, by testing many highly consistent Ta-doped Li7La3Zr2O12 (LLZTO) samples and utilizing an improved operando technique on symmetrical cells, we observed a consistent and statistically significant trend of current-dependent dendrite initiation time, which coincides with the Sand's time scaling in liquid electrolytes. This new understanding offers an electrochemical origin of dendritic initiation but also provides future insight into optimal all-solid-state battery design., Comment: 28 pages, 4 figures

This paper presents new methods that have the potential to improve privacy process efficiency with LLM and RAG. To reduce hallucination, we continually pre-train the base LLM model with a privacy-specific knowledge base and then augment it with a semantic RAG layer. Our evaluations demonstrate that this approach enhances the model performance (as much as doubled metrics compared to out-of-box LLM) in handling privacy-related queries, by grounding responses with factual information which reduces inaccuracies.

Lattice gauge theories (LGTs) can be employed to understand a wide range of phenomena, from elementary particle scattering in high-energy physics to effective descriptions of many-body interactions in materials. Studying dynamical properties of emergent phases can be challenging as it requires solving many-body problems that are generally beyond perturbative limits. We investigate the dynamics of local excitations in a $\mathbb{Z}_2$ LGT using a two-dimensional lattice of superconducting qubits. We first construct a simple variational circuit which prepares low-energy states that have a large overlap with the ground state; then we create particles with local gates and simulate their quantum dynamics via a discretized time evolution. As the effective magnetic field is increased, our measurements show signatures of transitioning from deconfined to confined dynamics. For confined excitations, the magnetic field induces a tension in the string connecting them. Our method allows us to experimentally image string dynamics in a (2+1)D LGT from which we uncover two distinct regimes inside the confining phase: for weak confinement the string fluctuates strongly in the transverse direction, while for strong confinement transverse fluctuations are effectively frozen. In addition, we demonstrate a resonance condition at which dynamical string breaking is facilitated. Our LGT implementation on a quantum processor presents a novel set of techniques for investigating emergent particle and string dynamics.

In this paper, we explore the impact of field redefinition on the spectrum of linearized perturbations in relativistic hydrodynamics. We observe that the spectrum of hydrodynamics modes is never affected by the local field redefinition, however, the spectrum of the non-hydrodynamic modes is affected. Through an appropriate all-order redefinition, non-hydrodynamic modes can be eliminated, leading to a new frame where the spectrum contains only hydrodynamic modes. We also observe that the resulting stress-energy tensor may have an infinite series in momentum space, with a convergence radius linked to the eliminated non-hydrodynamic mode. In certain special cases, higher-order terms in the stress-energy tensor under field redefinition may cancel, indicating that non-hydrodynamic modes are mere artefacts of the fluid variable choice and hold no physical significance, even if they appear to violate physical constraints. Using a special toy example, we find a criterion to distinguish between physical and unphysical non-hydrodynamic modes., Comment: 23 pages; comments and feedbacks are welcome; updated and extended; companion paper of arXiv:2410.19015

This paper improves upon existing data pruning methods for image classification by introducing a novel pruning metric and pruning procedure based on importance sampling. The proposed pruning metric explicitly accounts for data separability, data integrity, and model uncertainty, while the sampling procedure is adaptive to the pruning ratio and considers both intra-class and inter-class separation to further enhance the effectiveness of pruning. Furthermore, the sampling method can readily be applied to other pruning metrics to improve their performance. Overall, the proposed approach scales well to high pruning ratio and generalizes better across different classification models, as demonstrated by experiments on four benchmark datasets, including the fine-grained classification scenario.

Controlled transitions between coupling induced transparency (CIT) and coupling induced absorption (CIA) are effects of both fundamental importance as well as potential applications in various devices. We have explored these peculiar phenomena in multi-mode coupled hybrid quantum systems by considering a tunable mode (TM) and several static modes (SMs). The individual SMs and TM are designed such that they show CIA, but upon coupling different SMs we observe a transition from CIA to CIT. Quite remarkably we are able to achieve CIT using only purely dissipative couplings, whereas it is well known that CIT appears with coherent coupling. We have developed a robust quantum theory based formalism which is able to capture the transition between CIA to CIT and have the capability to explain the inter-transition (CIT to CIA) as well as intra-transitions (CIA to CIA, CIT to CIT etc.) in a multimode hybrid quantum system all with just linear approach. A general model is developed for hybrid quantum systems having N modes. We have then explicitly described two sets of hybrid systems, the first set is of three modes, 1TM coupled with 2SMs, and the second set is of four modes, 1TM coupled with 3SMs. Later we have generalised it for hybrid quantum systems having N number of modes. The results provide a pathway for designing hybrid systems that can control the group velocity of light, offering potential applications in the fields of optical switching and quantum information technology. Our finding and formulation that in a single hybrid quantum system we can achieve controllable inter-transitions and intra-transitions of CIT/ CIA may open a tool and guidance for its application in quantum technology and quantum materials as the TMs/SMs may be well extended to other real/ quasi-particles also.

Large Language Models (LLMs) have transformed natural language processing, yet improving their problem-solving capabilities, particularly for complex, reasoning-intensive tasks, remains a persistent challenge. This paper introduces the REAP (Reflection, Explicit Problem Deconstruction, and Advanced Prompting) method, an innovative approach within the dynamic context generation framework. REAP guides LLMs through reflection on the query, deconstructing it into manageable components, and generating relevant context to enhance the solution process. We evaluated REAP using a dataset designed to expose LLM limitations, comparing zero-shot prompting with REAP-enhanced prompts across six state-of-the-art models: OpenAI's o1-preview, o1-mini, GPT-4o, GPT-4o-mini, Google's Gemini 1.5 Pro, and Claude 3.5 Sonnet. The results demonstrate notable performance gains, with o1-mini improving by 40.97%, GPT-4o by 66.26%, and GPT-4o-mini by 112.93%. Despite the already strong baseline performance of OpenAI's o1-preview, modest gains were observed. Beyond performance improvements, REAP offers a cost-effective solution; for example, GPT-4o-mini, which is approximately 100 times cheaper than o1-preview, delivered competitive results. REAP also improves the clarity of model outputs, making it easier for humans to understand the reasoning behind the results and simplifying the process of identifying and addressing any issues. These findings demonstrate REAP's potential to greatly improve the capabilities of LLMs, providing both better performance and increased cost-efficiency across a wide range of applications., Comment: 524 pages, 3 figures

Ultra-low mass primordial black holes (PBH), briefly dominating the expansion of the universe, would leave detectable imprints in the secondary stochastic gravitational wave background (SGWB). Such a scenario leads to a characteristic doubly peaked spectrum of SGWB and strongly depends on the Hawking evaporation of such light PBHs. However, these observable signatures are significantly altered if the memory burden effect during the evaporation of PBHs is taken into account. We show that for the SGWB induced by PBH density fluctuations, the memory burden effects on the Hawking evaporation of ultra-low mass PBHs can mimic the signal arising due to the non-standard reheating epoch before PBH domination. Finally, we point out that this degeneracy can be broken by the simultaneous detection of the first peak in the SGWB, which is typically induced by the inflationary adiabatic perturbations., Comment: v1: 27 pages, 4 figures; v2: 28 pages, 4 figures, accepted for publication in JHEP

We report on the first experimental characterization of a laser-wakefield accelerator able to deliver, in a single pulse, doses in excess of 1 Gy on timescales of the order of tens of femtoseconds, reaching unprecedented average dose-rates above $10^{13}$ Gy/s. The irradiator is demonstrated to deliver doses tuneable up to 2.2 Gy in a cm$^2$ area and with a high degree of longitudinal and transverse uniformity in a single irradiation. In this regime, proof-of-principle irradiation of patient-derived glioblastoma stem-like cells and human skin fibroblast cells show indications of a differential cellular response, when compared to reference irradiations at conventional dose-rates. These include a statistically significant increase in relative biological effectiveness ($1.40\pm0.08$ at 50\% survival for both cell lines) and a significant reduction of the relative radioresistance of tumour cells. Data analysis provides preliminary indications that these effects might not be fully explained by induced oxygen depletion in the cells but may be instead linked to a higher complexity of the damages triggered by the ultra-high density of ionising tracks of femtosecond-scale radiation pulses. These results demonstrate an integrated platform for systematic radiobiological studies at unprecedented beam durations and dose-rates, a unique infrastructure for translational research in radiobiology at the femtosecond scale., Comment: 29 pages, 11 figures, Paper submission

Computational workflows are a common class of application on supercomputers, yet the loosely coupled and heterogeneous nature of workflows often fails to take full advantage of their capabilities. We created Colmena to leverage the massive parallelism of a supercomputer by using Artificial Intelligence (AI) to learn from and adapt a workflow as it executes. Colmena allows scientists to define how their application should respond to events (e.g., task completion) as a series of cooperative agents. In this paper, we describe the design of Colmena, the challenges we overcame while deploying applications on exascale systems, and the science workflows we have enhanced through interweaving AI. The scaling challenges we discuss include developing steering strategies that maximize node utilization, introducing data fabrics that reduce communication overhead of data-intensive tasks, and implementing workflow tasks that cache costly operations between invocations. These innovations coupled with a variety of application patterns accessible through our agent-based steering model have enabled science advances in chemistry, biophysics, and materials science using different types of AI. Our vision is that Colmena will spur creative solutions that harness AI across many domains of scientific computing.

Quantum error correction provides a path to reach practical quantum computing by combining multiple physical qubits into a logical qubit, where the logical error rate is suppressed exponentially as more qubits are added. However, this exponential suppression only occurs if the physical error rate is below a critical threshold. In this work, we present two surface code memories operating below this threshold: a distance-7 code and a distance-5 code integrated with a real-time decoder. The logical error rate of our larger quantum memory is suppressed by a factor of $\Lambda$ = 2.14 $\pm$ 0.02 when increasing the code distance by two, culminating in a 101-qubit distance-7 code with 0.143% $\pm$ 0.003% error per cycle of error correction. This logical memory is also beyond break-even, exceeding its best physical qubit's lifetime by a factor of 2.4 $\pm$ 0.3. We maintain below-threshold performance when decoding in real time, achieving an average decoder latency of 63 $\mu$s at distance-5 up to a million cycles, with a cycle time of 1.1 $\mu$s. To probe the limits of our error-correction performance, we run repetition codes up to distance-29 and find that logical performance is limited by rare correlated error events occurring approximately once every hour, or 3 $\times$ 10$^9$ cycles. Our results present device performance that, if scaled, could realize the operational requirements of large scale fault-tolerant quantum algorithms., Comment: 10 pages, 4 figures, Supplementary Information

This note provides a comprehensive examination of the various approaches to formulating relativistic hydrodynamics, with a particular emphasis on the canonical approach. Relativistic hydrodynamics plays a crucial role in understanding the behavior of fluids in high-energy astrophysical phenomena and heavy-ion collisions. The canonical approach is explored in detail, highlighting its foundational principles, mathematical formulations, and practical implications in modeling relativistic fluid dynamics. Following this, we delve into the linear response theory, elucidating its relevance in the context of hydrodynamics. We analyze the response of relativistic fluids to external perturbations, discussing the theoretical framework and key results. This dual focus aims to bridge the gap between theoretical foundations and practical applications, offering a robust perspective on the dynamic interplay between relativistic hydrodynamics and linear response theory.

Helium-burning stars, in particular Cepheids, are especially difficult to model, as the choice of free parameters can greatly impact the shape of the blue loops - the part of the evolutionary track at which instability strip is crossed. Contemporary one-dimensional stellar evolution codes, like MESA (Modules for Experiments in Stellar Astrophysics), come with a large number of free parameters that allow to model the physical processes in stellar interiors under many assumptions. The uncertainties that arise from this freedom are rarely discussed in the literature despite their impact on the evolution of the model. We calculate a grid of evolutionary models with MESA, varying several controls, like solar mixture of heavy elements, mixing length theory prescription, nuclear reaction rates, scheme to determine convective boundaries, atmosphere model, temporal and spatial resolution, and quantify their impact on age and location of the evolutionary track on the HR diagram from the main sequence till the end of core-helium burning. Our investigation was conducted for a full range of masses and metallicities expected for classical Cepheids. The uncertainties are significant, especially during core-helium burning, reaching or exceeding the observational uncertainties of log Teff and log L for detached eclipsing binary systems. For models below 9 solar masses, thin convective shells develop and evolve erratically, not allowing the models to converge. A careful inspection of Kippenhahn diagrams and convergence study is advised for a given mass and metallicity, to assess how severe this problem is and to what extent it may affect the evolution.

To address the data scarcity associated with 3D assets, 2D-lifting techniques such as Score Distillation Sampling (SDS) have become a widely adopted practice in text-to-3D generation pipelines. However, the diffusion models used in these techniques are prone to viewpoint bias and thus lead to geometric inconsistencies such as the Janus problem. To counter this, we introduce MT3D, a text-to-3D generative model that leverages a high-fidelity 3D object to overcome viewpoint bias and explicitly infuse geometric understanding into the generation pipeline. Firstly, we employ depth maps derived from a high-quality 3D model as control signals to guarantee that the generated 2D images preserve the fundamental shape and structure, thereby reducing the inherent viewpoint bias. Next, we utilize deep geometric moments to ensure geometric consistency in the 3D representation explicitly. By incorporating geometric details from a 3D asset, MT3D enables the creation of diverse and geometrically consistent objects, thereby improving the quality and usability of our 3D representations., Comment: 9 pages, 8 figures

Effective fraud detection and analysis of government-issued identity documents, such as passports, driver's licenses, and identity cards, are essential in thwarting identity theft and bolstering security on online platforms. The training of accurate fraud detection and analysis tools depends on the availability of extensive identity document datasets. However, current publicly available benchmark datasets for identity document analysis, including MIDV-500, MIDV-2020, and FMIDV, fall short in several respects: they offer a limited number of samples, cover insufficient varieties of fraud patterns, and seldom include alterations in critical personal identifying fields like portrait images, limiting their utility in training models capable of detecting realistic frauds while preserving privacy. In response to these shortcomings, our research introduces a new benchmark dataset, IDNet, designed to advance privacy-preserving fraud detection efforts. The IDNet dataset comprises 837,060 images of synthetically generated identity documents, totaling approximately 490 gigabytes, categorized into 20 types from $10$ U.S. states and 10 European countries. We evaluate the utility and present use cases of the dataset, illustrating how it can aid in training privacy-preserving fraud detection methods, facilitating the generation of camera and video capturing of identity documents, and testing schema unification and other identity document management functionalities., Comment: 40 pages

The fifth-generation new radio (5G NR) technology is expected to provide precise and reliable positioning capabilities along with high data rates. The Third Generation Partnership Project (3GPP) has started introducing positioning techniques from Release-16 based on time, angle, and signal strength using reference signals. However, validating these techniques with experimental prototypes is crucial before successful real-world deployment. This work provides useful tools and implementation details that are required in performing 5G positioning experiments with OpenAirInterface (OAI). As an example use case, we present an round trip time (RTT) estimation test-bed based on OAI and discusses the real-word experiment and measurement process.

The translocation of biopolymers, such as DNA and proteins, across cellular or nuclear membranes is essential for numerous biological processes. The translocation dynamics are influenced by the properties of the polymers, such as polymer stiffness, and the geometry of the capsid. In our study, we aim to investigate the impact of polymer stiffness, activity, and different capsid geometries on the packing and ejection dynamics of both passive and active polymers. We employ Langevin dynamics simulations for a systematic investigation. We observe that flexible polymers exhibit packing times that are faster than those of their semi-flexible counterparts. Interestingly, for large polymers compared to the capsid size, sphere facilitates faster packing and unpacking compared to ellipsoid, mimicking the cell nucleus and suggesting a geometrical advantage for biopolymer translocation. In summary, we observe that increasing activity accelerates both the packing and ejection processes for both flexible and semi-flexible polymers. However, the effect is significantly more pronounced for semi-flexible polymers, highlighting the crucial role of polymer flexibility in these dynamics. These findings deepen our understanding of the intricate interplay between polymer flexibility, capsid geometry, and activity, providing valuable insight into the dynamics of polymer packing and ejection processes., Comment: 10 pages, 9 figures

Background: The reproducibility of machine-learning models in prostate cancer detection across different MRI vendors remains a significant challenge. Methods: This study investigates Support Vector Machines (SVM) and Random Forest (RF) models trained on radiomic features extracted from T2-weighted MRI images using Pyradiomics and MRCradiomics libraries. Feature selection was performed using the maximum relevance minimum redundancy (MRMR) technique. We aimed to enhance clinical decision support through multimodal learning and feature fusion. Results: Our SVM model, utilizing combined features from Pyradiomics and MRCradiomics, achieved an AUC of 0.74 on the Multi-Improd dataset (Siemens scanner) but decreased to 0.60 on the Philips test set. The RF model showed similar trends, with notable robustness for models using Pyradiomics features alone (AUC of 0.78 on Philips). Conclusions: These findings demonstrate the potential of multimodal feature integration to improve the robustness and generalizability of machine-learning models for clinical decision support in prostate cancer detection. This study marks a significant step towards developing reliable AI-driven diagnostic tools that maintain efficacy across various imaging platforms.

Peculiar phenomena have been observed in analyses of anisotropic flow ($v_n$) fluctuations in ultracentral nucleus-nucleus collisions: The fourth-order cumulant of the elliptic flow ($v_2$) distribution changes sign. In addition, the ATLAS collaboration has shown that cumulants of $v_n$ fluctuations of all orders depend significantly on the centrality estimator. We show that these peculiarities are due to the fact that the impact parameter $b$ always spans a finite range for a fixed value of the centrality estimator. We provide a quantitative determination of this range through a simple Bayesian analysis. We obtain excellent fits of STAR and ATLAS data, with a few parameters, by assuming that the probability distribution of $v_n$ solely depends on $b$ at a given centrality. This probability distribution is almost Gaussian, and its parameters depend smoothly on $b$, in a way that is constrained by symmetry and scaling laws. We reconstruct, thus, the impact parameter dependence of the mean elliptic flow in the reaction plane in a model-independent manner, and assess the robustness of the extraction using Monte Carlo simulations of the collisions where the impact parameter is known. We argue that the non-Gaussianity of $v_n$ fluctuations gives direct information on the hydrodynamic response to initial anisotropies, ATLAS data being consistent with a smaller response for $n=4$ than for $n=2$ and $n=3$, in agreement with hydrodynamic calculations., Comment: 18 pages, 8 figures

We report on a direct search for elastic photon-photon scattering using x-ray and $\gamma$ photons from a laser-plasma based experiment. A gamma photon beam produced by a laser wakefield accelerator provided a broadband gamma spectrum extending to above $E_\gamma = 200$ MeV. These were collided with a dense x-ray field produced by the emission from a laser heated germanium foil at $E_x \approx 1.4$ keV, corresponding to an invariant mass of $\sqrt{s} = 1.22 \pm 0.22$ MeV. In these asymmetric collisions elastic scattering removes one x-ray and one high-energy $\gamma$ photon and outputs two lower energy $\gamma$ photons. No changes in the $\gamma$ photon spectrum were observed as a result of the collisions allowing us to place a 95% upper bound on the cross section of $1.5 \times 10^{15}\,\mu$b. Although far from the QED prediction, this represents the lowest upper limit obtained so far for $\sqrt{s} \lesssim 1$ MeV., Comment: 6 pages, 10 figures

In this work, we show that the kinetically constrained quantum East model lies between a quantum scarred and a many-body localized system featuring an unconventional type of mobility edge in the spectrum. We name this scenario $\textit{state-dependent}$ mobility edge: while the system does not exhibit a sharp separation in energy between thermal and non-thermal eigenstates, the abundance of non-thermal eigenstates results in slow entanglement growth for $\textit{many}$ initial states, such as product states, below a finite energy density. We characterize the state-dependent mobility edge by looking at the complexity of classically simulating dynamics using tensor network for system sizes well beyond those accessible via exact diagonalization. Focusing on initial product states, we observe a qualitative change in the dynamics of the bond dimension needed as a function of their energy density. Specifically, the bond dimension typically grows $\textit{polynomially}$ in time up to a certain energy density, where we locate the state-dependent mobility edge, enabling simulations for long times. Above this energy density, the bond dimension typically grows $\textit{exponentially}$ making the simulation practically unfeasible beyond short times, as generally expected in interacting theories. We correlate the polynomial growth of the bond dimension to the presence of many non-thermal eigenstates around that energy density, a subset of which we compute via tensor network., Comment: 12+3 pages; 8+5 figures

Understanding time-dependent effects of charged vacancies on electromechanical responses of materials is at the forefront of research for designing materials exhibiting metal-insulator transition, and memresistive behavior. A Rayleighian approach is used to develop a model for studying the non-linear kinetics of reaction leading to generation of vacancies and electrons by the dissociation of vacancy-electron pairs. Also, diffusion and elastic effects of charged vacancies are considered to model polarization-electric potential and strain-electric potential hysteresis loops. The model captures multi-physics phenomena by introducing couplings among polarization, electric potential, stress, strain, and concentrations of charged (multivalent) vacancies and electrons (treated as classical negatively charged particles), where the concentrations can vary due to association-dissociation reactions. Derivation of coupled time-dependent equations based on the Rayleighian approach is presented. Three limiting cases of the governing equations are considered highlighting effects of 1) non-linear reaction kinetics on the generation of charged vacancies and electrons, 2) the Vegard's law (i.e., the concentration-dependent local strain) on asymmetric strain-electric potential relations, and 3) coupling between a fast component and the slow component of the net polarization on the polarization-electric field relations. The Rayleighian approach discussed in this work should pave the way for developing a multi-scale modeling framework in a thermodynamically consistent manner while capturing multi-physics phenomena in ferroelectric materials.

Radiation reaction describes the effective force experienced by an accelerated charge due to radiation emission. Quantum effects dominate charge dynamics and radiation production[1][2] for charges accelerated by fields with strengths approaching the Schwinger field, $\mathbf{E_{sch}=}$\textbf{\SI[detect-weight]{1.3e18}{\volt\per\metre}[3]. Such fields exist in extreme astrophysical environments such as pulsar magnetospheres[4], may be accessed by high-power laser systems[5-7], dense particle beams interacting with plasma[8], crystals[9], and at the interaction point of next generation particle colliders[10]. Classical radiation reaction theories do not limit the frequency of radiation emitted by accelerating charges and omit stochastic effects inherent in photon emission[11], thus demanding a quantum treatment. Two quantum radiation reaction models, the quantum-continuous[12] and quantum-stochastic[13] models, correct the former issue, while only the quantum-stochastic model incorporates stochasticity[12]. Such models are of fundamental importance, providing insight into the effect of the electron self-force on its dynamics in electromagnetic fields. The difficulty of accessing conditions where quantum effects dominate inhibited previous efforts to observe quantum radiation reaction in charged particle dynamics with high significance. We report the first direct, high significance $(>5{\sigma})$ observation of strong-field radiation reaction on charged particles. Furthermore, we obtain strong evidence favouring the quantum radiation reaction models, which perform equivalently, over the classical model. Robust model comparison was facilitated by a novel Bayesian framework which inferred collision parameters. This framework has widespread utility for experiments where parameters governing lepton-laser collisions cannot be directly measured, including those using conventional accelerators., Comment: 31 pages, 14 figures

Heart rate (HR) is a crucial physiological signal that can be used to monitor health and fitness. Traditional methods for measuring HR require wearable devices, which can be inconvenient or uncomfortable, especially during sleep and meditation. Noncontact HR detection methods employing microwave radar can be a promising alternative. However, the existing approaches in the literature usually use high-gain antennas and require the sensor to face the user's chest or back, making them difficult to integrate into a portable device and unsuitable for sleep and meditation tracking applications. This study presents a novel approach for noncontact HR detection using a miniaturized Soli radar chip embedded in a portable device (Google Nest Hub). The chip has a $6.5 \mbox{ mm} \times 5 \mbox{ mm} \times 0.9 \mbox{ mm}$ dimension and can be easily integrated into various devices. The proposed approach utilizes advanced signal processing and machine learning techniques to extract HRs from radar signals. The approach is validated on a sleep dataset (62 users, 498 hours) and a meditation dataset (114 users, 1131 minutes). The approach achieves a mean absolute error (MAE) of $1.69$ bpm and a mean absolute percentage error (MAPE) of $2.67\%$ on the sleep dataset. On the meditation dataset, the approach achieves an MAE of $1.05$ bpm and a MAPE of $1.56\%$. The recall rates for the two datasets are $88.53\%$ and $98.16\%$, respectively. This study represents the first application of the noncontact HR detection technology to sleep and meditation tracking, offering a promising alternative to wearable devices for HR monitoring during sleep and meditation., Comment: 15 pages

Broadband absorption in the visible spectrum is essential in optoelectronic applications that involve power conversion such as photovoltaics and photocatalysis. Most ultrathin broadband absorbers use parasitic plasmonic structures that maximize absorption using surface plasmons and/or Fabry-Perot cavities, which limits the weight efficiency of the device. Here, we show the theoretical and experimental realization of an unpatterned/planar semiconductor thin-film absorber based on monolayer transition metal dichalcogenides (TMDCs). We experimentally demonstrate an average total absorption in the visible range (450 nm - 700 nm) of > 70% using > 4 nm of semiconductor absorbing materials scalable over large areas with vapor phase growth techniques. Our analysis suggests that a power conversion efficiency (PCE) of 15.54% and a specific power > 300 W g^-1 may be achieved in a photovoltaic cell based on this metamaterial absorber.

This demo paper presents a dApp-based real-time spectrum sharing scenario where a 5th generation (5G) base station implementing the NR stack adapts its transmission and reception strategies based on the incumbent priority users in the Citizen Broadband Radio Service (CBRS) band. The dApp is responsible for obtaining relevant measurements from the Next Generation Node Base (gNB), running the spectrum sensing inference, and configuring the gNB with a control action upon detecting the primary incumbent user transmissions. This approach is built on dApps, which extend the O-RAN framework to the real-time and user plane domains. Thus, it avoids the need of dedicated Spectrum Access Systems (SASs) in the CBRS band. The demonstration setup is based on the open-source 5G OpenAirInterface (OAI) framework, where we have implemented a dApp interfaced with a gNB and communicating with a Commercial Off-the-Shelf (COTS) User Equipment (UE) in an over-the-air wireless environment. When an incumbent user has active transmission, the dApp will detect and inform the primary user presence to the gNB. The dApps will also enforce a control policy that adapts the scheduling and transmission policy of the Radio Access Network (RAN). This demo provides valuable insights into the potential of using dApp-based spectrum sensing with O-RAN architecture in next generation cellular networks.

Here, we study YbI$_3$, a quasi-2D layered material with Yb atoms arranged on an ideal honeycomb network of edge-sharing YbI$_6$ octahedra, analogous to the low-temperature phase of $\alpha-$RuCl$_3$. High quality single crystals of YbI$_3$ are grown from Yb and I as starting precursors, using the vapor transport technique. The grown crystals are characterized by single crystal x-ray diffraction, Raman spectroscopy, magnetization, and heat capacity probes. The crystal-field split ground state of Yb$^{3+}$ in \Yb~ is a well-isolated Kramers doublet with an effective moment $\rm J_{eff} = 1/2$. Upon cooling, the low-temperature heat capacity of \Yb~ reveals a broad peak at $\rm T_1 = 0.95$~K due to short-range ordering of the Yb moments, followed by a sharp peak at $\rm T_2 = T_N = 0.6$~K due to long-range ordering. The magnetic behavior is found to be weakly anisotropic with $\chi^\parallel > \chi^\perp$, where $\chi^\parallel$ and $\chi^\perp$ refers to the in-plane ($H \parallel ab$) and out-of-plane ($H \perp ab$) susceptibilities. The 2~K isothermal magnetization saturates at $\rm \approx~1.5~\mu_B/Yb^{3+}$ (in-plane) and $\rm \approx~1~\mu_B/Yb^{3+}$ (out-of-plane), suggesting the anisotropy to be easy-plane type. Low-temperature heat capacity, well below T$_N$, is found to vary as T$^\alpha$ with $\alpha~\approx~2.5$, indicating a possible unconventional magnetic ground state for YbI$_3$., Comment: 10 pages, 5 figures, Supplemental Material available on request

Software developers commonly rely on platforms like Stack Overflow for problem-solving and learning. However, academic research is an untapped resource that could greatly benefit industry practitioners. The challenge lies in connecting the innovative insights from academia to real-world problems faced by developers. This project introduces ResearchBot, a tool designed to bridge this academia-industry gap. ResearchBot employs a modular approach, encompassing understanding questions, curating queries to obtain relevant papers in the CrossRef repository, summarizing paper content and finally answering user questions based on paper summaries. The core objective of ResearchBot is to democratize access to academic knowledge for industry professionals. By providing concise summaries of cutting-edge research directly in response to SE-related questions, ResearchBot facilitates the application of academic insights to practical contexts. Ultimately, it aims to bridge the gap between academia and industry, using research evidence to support learning and decision-making in software development.

This study presents a computer vision approach aimed at detecting snow on sidewalks and pavements to reduce winter-related fall injuries, especially among elderly and visually impaired individuals. Leveraging fine-tuned VGG-19 and ResNet50 convolutional neural networks (CNNs), the research focuses on identifying snow presence in pavement images. The dataset comprises 98 images evenly split between snowy and snow-free conditions, evaluated with a separate test set using the F1 score and accuracy metrics. This work builds upon existing research by employing fine-tuned CNN architectures to accurately detect snow on pavements from smartphone-captured images. The methodology incorporates transfer learning and model ensembling techniques to integrate the best predictions from both the VGG19 and ResNet50 architectures. The study yields accuracy and F1 scores of 81.8% and 81.7%, respectively, showcasing the potential of computer vision in addressing winter-related hazards for vulnerable populations., Comment: 5 pages 1 figure 1 table Included in MWAIS - 2024 Conference Proceedings Chair: Jacob Young

This study presents the theoretical foundations of an analogous electromagnetically induced transparency (EIT) and absorption (EIA) which we are referring as coupling induced transparency (CIT) and absorption (CIA) respectively, along with an exploration of the transition between these phenomena. We provide a concise phenomenological description with analytical expressions for transmission spectra and dispersion elucidating how the interplay of coherent and dissipative interactions in a coupled system results in the emergence of level repulsion and attraction, corresponding to CIT and CIA, respectively. The model is validated through numerical simulations using a hybrid system comprising a split ring resonator (SRR) and electric inductive-capacitive (ELC) resonator in planar geometry. We analyse two cases while keeping ELC parameters constant; one involving a dynamic adjustment of the SRR size with a fixed split gap, and the other entailing a varying gap while maintaining a constant SRR size. Notably, in the first case, the dispersion profile of the transmission signal demonstrates level repulsion, while the second case results in level attraction, effectively showcasing CIT and CIA, respectively. These simulated findings not only align with the theoretical model but also underscore the versatility of our approach. Subsequently, we expand our model to a more general case, demonstrating that a controlled transition from CIT to CIA is achievable by manipulating the dissipation rate of individual modes within the hybrid system, leading to either coherent or dissipative interactions between the modes. The results provide a pathway for designing hybrid systems that can control the group velocity of light, offering potential applications in the fields of optical switching and quantum information technology.

In this article, we define Dirichlet-type space $\mathcal{D}^{2}(\boldsymbol{\mu})$ over the bidisc $\mathbb D^2$ for any measure $\boldsymbol{\mu}\in\mathcal{P}\mathcal{M}_{+}(\mathbb T^2).$ We show that the set of polynomials is dense in $\mathcal{D}^{2}(\boldsymbol{\mu})$ and the pair $(M_{z_1}, M_{z_2})$ of multiplication operator by co-ordinate functions on $\mathcal{D}^{2}(\boldsymbol{\mu})$ is a pair of commuting $2$-isometries. Moreover, the pair $(M_{z_1}, M_{z_2})$ is a left-inverse commuting pair in the following sense: $L_{M_{z_i}} M_{z_j}=M_{z_j}L_{M_{z_i}}$ for $1\leqslant i\neq j\leqslant n,$ where $L_{M_{z_i}}$ is the left inverse of $M_{z_i}$ with $\ker L_{M_{z_i}} =\ker M_{z_i}^*$, $1\leqslant i \leqslant n$. Furthermore, it turns out that, for the class of left-inverse commuting tuple $\boldsymbol T=(T_1, \ldots, T_n)$ acting on a Hilbert space $\mathcal{H}$, the joint wandering subspace property is equivalent to the individual wandering subspace property. As an application of this, the article shows that the class of left-inverse commuting pair with certain splitting property is modelled by the pair of multiplication by co-ordinate functions $(M_{z_1}, M_{z_2})$ on $\mathcal{D}^{2}(\boldsymbol{\mu})$ for some $\boldsymbol{\mu}\in\mathcal{P}\mathcal{M}_{+}(\mathbb T^2).$, Comment: The article has been significantly revised

We study how to subvert large language models (LLMs) from following prompt-specified rules. We model rule-following as inference in propositional Horn logic, a mathematical system in which rules have the form ``if $P$ and $Q$, then $R$'' for some propositions $P$, $Q$, and $R$. We prove that although LLMs can faithfully follow such rules, maliciously crafted prompts can mislead even idealized, theoretically constructed models. Empirically, we find that the reasoning behavior of LLMs aligns with that of our theoretical constructions, and popular attack algorithms find adversarial prompts with characteristics predicted by our theory. Our logic-based framework provides a novel perspective for mechanistically understanding the behavior of LLMs in rule-based settings such as jailbreak attacks.

In this paper, we propose a novel token selective attention approach, ToSA, which can identify tokens that need to be attended as well as those that can skip a transformer layer. More specifically, a token selector parses the current attention maps and predicts the attention maps for the next layer, which are then used to select the important tokens that should participate in the attention operation. The remaining tokens simply bypass the next layer and are concatenated with the attended ones to re-form a complete set of tokens. In this way, we reduce the quadratic computation and memory costs as fewer tokens participate in self-attention while maintaining the features for all the image patches throughout the network, which allows it to be used for dense prediction tasks. Our experiments show that by applying ToSA, we can significantly reduce computation costs while maintaining accuracy on the ImageNet classification benchmark. Furthermore, we evaluate on the dense prediction task of monocular depth estimation on NYU Depth V2, and show that we can achieve similar depth prediction accuracy using a considerably lighter backbone with ToSA., Comment: Accepted at CVPRW 2024

Many computational tasks can be naturally expressed as a composition of a DNN followed by a program written in a traditional programming language or an API call to an LLM. We call such composites "neural programs" and focus on the problem of learning the DNN parameters when the training data consist of end-to-end input-output labels for the composite. When the program is written in a differentiable logic programming language, techniques from neurosymbolic learning are applicable, but in general, the learning for neural programs requires estimating the gradients of black-box components. We present an algorithm for learning neural programs, called ISED, that only relies on input-output samples of black-box components. For evaluation, we introduce new benchmarks that involve calls to modern LLMs such as GPT-4 and also consider benchmarks from the neurosymbolic learning literature. Our evaluation shows that for the latter benchmarks, ISED has comparable performance to state-of-the-art neurosymbolic frameworks. For the former, we use adaptations of prior work on gradient approximations of black-box components as a baseline, and show that ISED achieves comparable accuracy but in a more data- and sample-efficient manner.

The macroscopic energy-momentum and spin densities of relativistic spin hydrodynamics are obtained from the ensemble average of their respective microscopic definitions (quantum operators). These microscopic definitions suffer from ambiguities, meaning that, one may obtain different forms of symmetric energy-momentum tensor and spin tensor through pseudogauge transformations (or in other words Belinfante improvement procedure). However, this ambiguity may be removed if we obtain these currents using Noether's second theorem instead of widely used Noether's first theorem. In this article, we use Noether's second theorem to derive energy-momentum and spin currents without the need of pseudogauge transformations for free Dirac massive particles with spin one-half., Comment: Updated

This paper introduces Fortune Analytics Language Model (FALM). FALM empowers users with direct access to comprehensive business analysis, including market trends, company performance metrics, and expert insights. Unlike generic LLMs, FALM leverages a curated knowledge base built from professional journalism, enabling it to deliver precise and in-depth answers to intricate business questions. Users can further leverage natural language queries to directly visualize financial data, generating insightful charts and graphs to understand trends across diverse business sectors clearly. FALM fosters user trust and ensures output accuracy through three novel methods: 1) Time-aware reasoning guarantees accurate event registration and prioritizes recent updates. 2) Thematic trend analysis explicitly examines topic evolution over time, providing insights into emerging business landscapes. 3) Content referencing and task decomposition enhance answer fidelity and data visualization accuracy. We conduct both automated and human evaluations, demonstrating FALM's significant performance improvements over baseline methods while prioritizing responsible AI practices. These benchmarks establish FALM as a cutting-edge LLM in the business and media domains, with exceptional accuracy and trustworthiness.

The flow-induced vibration and cavitation dynamics of three-dimensional flow past a cantilever flexible hydrofoil are investigated using a large eddy simulation (LES) model, a homogeneous mixture cavitation model and the structural modes superposition method. The present work aims to explore a potential mechanism responsible for a propeller singing behavior, and thus focuses on the synchronized hydroelastic coupling among the pressure pulsation inside the flow field, the cavitation generation and the structural vibration. To begin, we validate the tip vortex dynamics of a flexible hydrofoil against the available experimental. Our results demonstrate that the tip vortex shedding and the blade vibration are responsible for the intense peak in the low-frequency tonal components of the noise source, and the trailing-edge vortex shedding induces broadband components. Additionally, the generation of sheet cavitation induces considerable synchronized hydrofoil vibration (subjected to a flutter-like response), and affects the pressure fluctuations in the flow field, which further dominate the features of the underwater noise sources. It is suggested that the cavitation behavior and structural vibrations co-dominate the characteristics of singing noise from a propeller blade.