Search

Learning conditional distributions $\pi^*(\cdot|x)$ is a central problem in machine learning, which is typically approached via supervised methods with paired data $(x,y) \sim \pi^*$. However, acquiring paired data samples is often challenging, especially in problems such as domain translation. This necessitates the development of $\textit{semi-supervised}$ models that utilize both limited paired data and additional unpaired i.i.d. samples $x \sim \pi^*_x$ and $y \sim \pi^*_y$ from the marginal distributions. The usage of such combined data is complex and often relies on heuristic approaches. To tackle this issue, we propose a new learning paradigm that integrates both paired and unpaired data $\textbf{seamlessly}$ through the data likelihood maximization techniques. We demonstrate that our approach also connects intriguingly with inverse entropic optimal transport (OT). This finding allows us to apply recent advances in computational OT to establish a $\textbf{light}$ learning algorithm to get $\pi^*(\cdot|x)$. Furthermore, we demonstrate through empirical tests that our method effectively learns conditional distributions using paired and unpaired data simultaneously.

The Iterative Markovian Fitting (IMF) procedure based on iterative reciprocal and Markovian projections has recently been proposed as a powerful method for solving the Schr\"odinger Bridge problem. However, it has been observed that for the practical implementation of this procedure, it is crucial to alternate between fitting a forward and backward time diffusion at each iteration. Such implementation is thought to be a practical heuristic, which is required to stabilize training and obtain good results in applications such as unpaired domain translation. In our work, we show that this heuristic closely connects with the pioneer approaches for the Schr\"odinger Bridge based on the Iterative Proportional Fitting (IPF) procedure. Namely, we find that the practical implementation of IMF is, in fact, a combination of IMF and IPF procedures, and we call this combination the Iterative Proportional Markovian Fitting (IPMF) procedure. We show both theoretically and practically that this combined IPMF procedure can converge under more general settings, thus, showing that the IPMF procedure opens a door towards developing a unified framework for solving Schr\"odinger Bridge problems.

Understanding how features evolve across layers in deep neural networks is a fundamental challenge in mechanistic interpretability, particularly due to polysemanticity and feature superposition. While Sparse Autoencoders (SAEs) have been used to extract interpretable features from individual layers, aligning these features across layers has remained an open problem. In this paper, we introduce SAE Match, a novel, data-free method for aligning SAE features across different layers of a neural network. Our approach involves matching features by minimizing the mean squared error between the folded parameters of SAEs, a technique that incorporates activation thresholds into the encoder and decoder weights to account for differences in feature scales. Through extensive experiments on the Gemma 2 language model, we demonstrate that our method effectively captures feature evolution across layers, improving feature matching quality. We also show that features persist over several layers and that our approach can approximate hidden states across layers. Our work advances the understanding of feature dynamics in neural networks and provides a new tool for mechanistic interpretability studies.

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.

ThMn12-type (Sm,Zr)1(Fe,Co,Ti)12 compounds show great potential for permanent magnets. Magnetically hard anisotropic powders prepared via reduction-diffusion exhibit a significant increase in coercivity from 0.45 T to 1.26 T as the processing temperature is raised from 990{\deg}C to 1220{\deg}C. Structural and microchemical analyses at high-resolution reveal that high-temperature processing annihilates grain boundaries (GBs) and reduces the density of twin boundaries (TBs), which are defects acting as weak links limiting the coercivity in the 1:12 system. Ostwald ripening is proposed as the mechanism behind the reduction of GB and TB densities at higher temperature, driven by the reduction in interfacial energy and enhancing atomic structural uniformity.

This approach is based on the idea used to construct the so-called ${\bf fr}$-language. We construct such functorial languages as the ${\bf fr_{\infty}}$-language and other versions of it on categories of groups using the lower central series of groups of relations. In addition, we compute higher limits of the coinvariants of exterior powers, free Lie powers of modules of relations associated to free representations of groups. We show that such ${\bf fr_{\infty}}$-languages have remarkable properties when restricted to groups of finite homological dimension, torsion-free groups, and groups having no torsion up to $n$. Finally, we show how one can use functorial languages to transform categories/varieties of groups into certain spectra, which we call flux-spectra.

This paper introduces Idempotent Test-Time Training (IT$^3$), a novel approach to addressing the challenge of distribution shift. While supervised-learning methods assume matching train and test distributions, this is rarely the case for machine learning systems deployed in the real world. Test-Time Training (TTT) approaches address this by adapting models during inference, but they are limited by a domain specific auxiliary task. IT$^3$ is based on the universal property of idempotence. An idempotent operator is one that can be applied sequentially without changing the result beyond the initial application, that is $f(f(x))=f(x)$. At training, the model receives an input $x$ along with another signal that can either be the ground truth label $y$ or a neutral "don't know" signal $0$. At test time, the additional signal can only be $0$. When sequentially applying the model, first predicting $y_0 = f(x, 0)$ and then $y_1 = f(x, y_0)$, the distance between $y_0$ and $y_1$ measures certainty and indicates out-of-distribution input $x$ if high. We use this distance, that can be expressed as $||f(x, f(x, 0)) - f(x, 0)||$ as our TTT loss during inference. By carefully optimizing this objective, we effectively train $f(x,\cdot)$ to be idempotent, projecting the internal representation of the input onto the training distribution. We demonstrate the versatility of our approach across various tasks, including corrupted image classification, aerodynamic predictions, tabular data with missing information, age prediction from face, and large-scale aerial photo segmentation. Moreover, these tasks span different architectures such as MLPs, CNNs, and GNNs.

Federated Learning (FL) facilitates data privacy by enabling collaborative in-situ training across decentralized clients. Despite its inherent advantages, FL faces significant challenges of performance and convergence when dealing with data that is not independently and identically distributed (non-i.i.d.). While previous research has primarily addressed the issue of skewed label distribution across clients, this study focuses on the less explored challenge of multi-domain FL, where client data originates from distinct domains with varying feature distributions. We introduce a novel method designed to address these challenges FedStein: Enhancing Multi-Domain Federated Learning Through the James-Stein Estimator. FedStein uniquely shares only the James-Stein (JS) estimates of batch normalization (BN) statistics across clients, while maintaining local BN parameters. The non-BN layer parameters are exchanged via standard FL techniques. Extensive experiments conducted across three datasets and multiple models demonstrate that FedStein surpasses existing methods such as FedAvg and FedBN, with accuracy improvements exceeding 14% in certain domains leading to enhanced domain generalization. The code is available at https://github.com/sunnyinAI/FedStein, Comment: 12 pages, 2 figures. Accepted at International Workshop on Federated Foundation Models In Conjunction with NeurIPS 2024 (FL@FM-NeurIPS'24)

Inspired by the recent success of machine-learned interatomic potentials for crystal structure prediction of the inorganic crystals, we present a methodology that exploits Moment Tensor Potentials and active learning (based on maxvol algorithm) to accelerate structure prediction of molecular crystals. Benzene and glycine are used as test systems. Interestingly, among obtained low energy structures of benzene we have found a peculiar polymeric benzene structure.

Photon entanglement is indispensable for optical quantum technologies. Measurement-based optical quantum computing and all-optical quantum networks rely on multiphoton cluster states consisting of indistinguishable entangled photons. A promising method for creating such cluster states on demand is spin-photon entanglement using the spin of a resident charge carrier in a quantum dot, precessing in a weak external magnetic field. In this work, we show theoretically and experimentally that spin-photon entanglement is strongly affected by the hidden anisotropy of quantum dots, which can arise from mechanical stress, shape anisotropy and even specific crystal structure. In the measurements of time-resolved photoluminescence and cross-polarized second-order photon correlation function in a magnetic field, the anisotropy manifests itself in the spin dynamics and, as a consequence, in the spin-photon concurrence. The measured time-filtered spin-photon Bell state fidelity depends strongly on the excitation polarization and reaches an extremely high value of 94% at maximum. We specify the magnetic field and excitation polarization directions that maximize spin-photon entanglement and thereby enhance the fidelity of multiphoton entangled states.

We give an explicit counterexample to the Bunkbed Conjecture introduced by Kasteleyn in 1985. The counterexample is given by a planar graph on $7222$ vertices, and is built on the recent work of Hollom (2024)., Comment: 12 pages, 2 figures

In this work, we describe our submissions for the Voice Privacy Challenge 2024. Rather than proposing a novel speech anonymization system, we enhance the provided baselines to meet all required conditions and improve evaluated metrics. Specifically, we implement emotion embedding and experiment with WavLM and ECAPA2 speaker embedders for the B3 baseline. Additionally, we compare different speaker and prosody anonymization techniques. Furthermore, we introduce Mean Reversion F0 for B5, which helps to enhance privacy without a loss in utility. Finally, we explore disentanglement models, namely $\beta$-VAE and NaturalSpeech3 FACodec., Comment: System description for VPC 2024

We define an index for invertible phases of two-dimensional fermionic and bosonic quantum spin systems without any additional symmetry. Conjecturally, it provides a microscopic definition of an invariant related to the chiral central charge of the boundary modes $c_- \bmod 24$ when the effective conformal field theory description is valid. Using this index, we prove that free fermionic systems with Chern number $\nu \bmod 48 \neq 0$ are in a non-trivial invertible phase., Comment: 21 pages, 6 figures

Background: The COVID-19 pandemic has highlighted the need for robust diagnostic tools capable of detecting the disease from diverse and evolving data sources. Machine learning models, especially convolutional neural networks (CNNs), have shown promise. However, the dynamic nature of real-world data can lead to model drift, where performance degrades over time as the underlying data distribution changes. Addressing this challenge is crucial to maintaining accuracy and reliability in diagnostic applications. Objective: This study aims to develop a framework that monitors model drift and employs adaptation mechanisms to mitigate performance fluctuations in COVID-19 detection models trained on dynamic audio data. Methods: Two crowd-sourced COVID-19 audio datasets, COVID-19 Sounds and COSWARA, were used. Each was divided into development and post-development periods. A baseline CNN model was trained and evaluated using cough recordings from the development period. Maximum mean discrepancy (MMD) was used to detect changes in data distributions and model performance between periods. Upon detecting drift, retraining was triggered to update the baseline model. Two adaptation approaches were compared: unsupervised domain adaptation (UDA) and active learning (AL). Results: UDA improved balanced accuracy by up to 22% and 24% for the COVID-19 Sounds and COSWARA datasets, respectively. AL yielded even greater improvements, with increases of up to 30% and 60%, respectively. Conclusions: The proposed framework addresses model drift in COVID-19 detection, enabling continuous adaptation to evolving data. This approach ensures sustained model performance, contributing to robust diagnostic tools for COVID-19 and potentially other infectious diseases.

Multi-distribution or collaborative learning involves learning a single predictor that works well across multiple data distributions, using samples from each during training. Recent research on multi-distribution learning, focusing on binary loss and finite VC dimension classes, has shown near-optimal sample complexity that is achieved with oracle efficient algorithms. That is, these algorithms are computationally efficient given an efficient ERM for the class. Unlike in classical PAC learning, where the optimal sample complexity is achieved with deterministic predictors, current multi-distribution learning algorithms output randomized predictors. This raises the question: can these algorithms be derandomized to produce a deterministic predictor for multiple distributions? Through a reduction to discrepancy minimization, we show that derandomizing multi-distribution learning is computationally hard, even when ERM is computationally efficient. On the positive side, we identify a structural condition enabling an efficient black-box reduction, converting existing randomized multi-distribution predictors into deterministic ones.

The principal component of conventional database query optimizers is a cost model that is used to estimate expected performance of query plans. The accuracy of the cost model has direct impact on the optimality of execution plans selected by the optimizer and thus, on the resulting query latency. Several common parameters of cost models in modern DBMS are related to the performance of CPU and I/O and are typically set by a database administrator upon system tuning. However these performance characteristics are not stable and therefore, a single point estimation may not suffice for all DB load regimes. In this paper, we propose an Adaptive Cost Model (ACM) which dynamically optimizes CPU- and I/O-related plan cost parameters at DB runtime. By continuously monitoring query execution statistics and the state of DB buffer cache ACM adjusts cost parameters without the need for manual intervention from a database administrator. This allows for responding to changes in the workload and system performance ensuring more optimal query execution plans. We describe the main ideas in the implementation of ACM and report on a preliminary experimental evaluation showing 20\% end-to-end latency improvement on TPC-H benchmark.

Fast Radio Bursts (FRBs) are millisecond-duration events detected from beyond the Milky Way. FRB emission characteristics favor highly magnetized neutron stars, or magnetars, as the sources, as evidenced by FRB-like bursts from a galactic magnetar, and the star-forming nature of FRB host galaxies. However, the processes that produce FRB sources remain unknown. Although galactic magnetars are often linked to core-collapse supernovae (CCSNe), it's uncertain what determines which supernovae result in magnetars. The galactic environments of FRB sources can be harnessed to probe their progenitors. Here, we present the stellar population properties of 30 FRB host galaxies discovered by the Deep Synoptic Array. Our analysis shows a significant deficit of low-mass FRB hosts compared to the occurrence of star-formation in the universe, implying that FRBs are a biased tracer of star-formation, preferentially selecting massive star-forming galaxies. This bias may be driven by galaxy metallicity, which is positively correlated with stellar mass. Metal-rich environments may favor the formation of magnetar progenitors through stellar mergers, as higher metallicity stars are less compact and more likely to fill their Roche lobes, leading to unstable mass transfer. Although massive stars do not have convective interiors to generate strong magnetic fields by dynamo, merger remnants are thought to have the requisite internal magnetic-field strengths to result in magnetars. The preferential occurrence of FRBs in massive star-forming galaxies suggests that CCSN of merger remnants preferentially forms magnetars., Comment: Accepted for publication in Nature. The final version will be published by the journal

Approximately half of the Universe's dark matter resides in collapsed halos; significantly less than half of the baryonic matter (protons and neutrons) remains confined to halos. A small fraction of baryons are in stars and the interstellar medium within galaxies. The lion's share are diffuse (less than $10^{-3}$ cm$^{-3}$) and ionized (neutral fraction less than $10^{-4}$), located in the intergalactic medium (IGM) and in the halos of galaxy clusters, groups, and galaxies. The quantity and spatial distribution of this diffuse ionized gas is notoriously difficult to measure, but has wide implications for galaxy formation, astrophysical feedback, and precision cosmology. Recently, the dispersion of extragalactic Fast Radio Bursts (FRBs) has been used to measure the total content of cosmic baryons. However, past efforts had modest samples and methods that cannot discriminate between IGM and halo gas, which is critical for studying feedback and for observational cosmology. Here, we present a large cosmological sample of FRB sources localized to their host galaxies. We have robustly partitioned the missing baryons into the IGM, galaxy clusters, and galaxies, providing a late-Universe measurement of the total baryon density of $\Omega_b h_{70}$=0.049$\pm$0.003. Our results indicate efficient feedback processes that can expel gas from galaxy halos and into the intergalactic medium, agreeing with the enriched cosmic web scenario seen in cosmological simulations. The large diffuse baryon fraction that we have measured disfavours bottom-heavy stellar initial mass functions, which predict a large total stellar density, $\Omega_*$.

We present our findings from the first long X-ray observation of the hard X-ray source Swift J0826.2-7033 with XMM-Newton, which has shown characteristics of magnetic accretion. The system appears to have a long orbital period (~7.8 h) accompanied by short timescale variabilities, which we tentatively interpret as the spin and beat periods of an intermediate polar. These short- and long-timescale modulations are energy-independent, suggesting that photoelectric absorption does not play any role in producing the variabilities. If our suspected spin and beat periods are true, then Swift J0826.2-7033 accretes via disc-overflow with an equal fraction of accretion taking place via disc and stream. The XMM-Newton and Swift-BAT spectral analysis reveals that the post-shock region in Swift J0826.2-7033 has a multi-temperature structure with a maximum temperature of ~43 keV, which is absorbed by a material with an average equivalent hydrogen column density of ~1.6$\times$10$^{22}$ cm$^{-2}$ that partially covers ~27% of the X-ray source. The suprasolar abundances, with hints of an evolved donor, collectively make Swift J0826.2-7033 an interesting target, which likely underwent a thermal timescale mass transfer phase., Comment: 10 pages, 7 figures. Accepted for publication in A&A

Relativistic outflows emanating from active galactic nuclei can extend up to kiloparsec scales in length, displaying a variety of complex morphologies. This study explores the intricate morphologies of such relativistic jets, mainly focusing on creating a bridge between magnetic instabilities in jets with observational signatures from complex radio galaxies. In particular, we aim to study the role of dynamical instabilities in forming distinctive morphological features by employing 3D relativistic magnetohydrodynamic (RMHD) simulations of rotating jets. Our simulations have further used the hybrid Eulerian-Lagrangian framework of the PLUTO code and generated the synthetic synchrotron emission and polarisation maps to compare with the observed signatures. Our analysis based on simulations of a continuously injected jet suggests that current-driven instabilities, notably the $|m|=1$ mode, generate rib-like structures that are seen in some of the recent radio galaxies using MeerKat, e.g. MysTail. In our contrasting simulations of the restarted jet, the kink-instability driven rib-like structures were formed relatively near the nozzle. In both cases, the jet dissipates its pre-existing magnetic energy through these instabilities, transitioning to a more kinetic energy dominant state. The turbulent structures resulting from this dissipation phase are filamentary and resemble the tethers as observed for the case of MysTail. This pilot study essentially provides a plausible qualitative explanation by bridging simulations of kink instability to produce synthetic radio features resembling the observed complex radio morphology of MysTail., Comment: 28 Pages, 18 Figures, 1 Table, Accepted for publication in Journal of High Energy Astrophysics

Microstructure plays a critical role in determining the macroscopic properties of materials, with applications spanning alloy design, MEMS devices, and tissue engineering, among many others. Computational frameworks have been developed to capture the complex relationship between microstructure and material behavior. However, despite these advancements, the steep learning curve associated with domain-specific knowledge and complex algorithms restricts the broader application of these tools. To lower this barrier, we propose a framework that integrates Natural Language Processing (NLP), Large Language Models (LLMs), and Denoising Diffusion Probabilistic Models (DDPMs) to enable microstructure design using intuitive natural language commands. Our framework employs contextual data augmentation, driven by a pretrained LLM, to generate and expand a diverse dataset of microstructure descriptors. A retrained NER model extracts relevant microstructure descriptors from user-provided natural language inputs, which are then used by the DDPM to generate microstructures with targeted mechanical properties and topological features. The NLP and DDPM components of the framework are modular, allowing for separate training and validation, which ensures flexibility in adapting the framework to different datasets and use cases. A surrogate model system is employed to rank and filter generated samples based on their alignment with target properties. Demonstrated on a database of nonlinear hyperelastic microstructures, this framework serves as a prototype for accessible inverse design of microstructures, starting from intuitive natural language commands., Comment: 29 pages, 15 figures

We compute the co-multiplication of the algebraic Morava K-theory for split orthogonal groups. This allows us to compute the decomposition of the Morava motives of generic maximal orthogonal Grassmannians and to compute a Morava K-theory analogue of the $J$-invariant in terms of the ordinary (Chow) $J$-invariant.

The loss landscape of neural networks is a critical aspect of their training, and understanding its properties is essential for improving their performance. In this paper, we investigate how the loss surface changes when the sample size increases, a previously unexplored issue. We theoretically analyze the convergence of the loss landscape in a fully connected neural network and derive upper bounds for the difference in loss function values when adding a new object to the sample. Our empirical study confirms these results on various datasets, demonstrating the convergence of the loss function surface for image classification tasks. Our findings provide insights into the local geometry of neural loss landscapes and have implications for the development of sample size determination techniques.

Consensus plays a crucial role in distributed ledger systems, impacting both scalability and decentralization. Many blockchain systems use a weighted lottery based on a scarce resource such as a stake, storage, memory, or computing power to select a committee whose members drive the consensus and are responsible for adding new information to the ledger. Therefore, ensuring a robust and fair committee selection process is essential for maintaining security, efficiency, and decentralization. There are two main approaches to randomized committee selection. In one approach, each validator candidate locally checks whether they are elected to the committee and reveals their proof during the consensus phase. In contrast, in the second approach, a sortition algorithm decides a fixed-sized committee that is globally verified. This paper focuses on the latter approach, with cryptographic sortition as a method for fair committee selection that guarantees a constant committee size. Our goal is to develop deterministic guarantees that strengthen decentralization. We introduce novel methods that provide deterministic bounds on the influence of adversaries within the committee, as evidenced by numerical experiments. This approach overcomes the limitations of existing protocols that only offer probabilistic guarantees, often providing large committees that are impractical for many quorum-based applications like atomic broadcast and randomness beacon protocols.

New formulae for a summation over a positive part of $SL(2,\mathbb Z)$ are presented. Such formulae can be written for any convex curve. We present several formulae where $\pi$ is obtained.

The SoccerNet 2024 challenges represent the fourth annual video understanding challenges organized by the SoccerNet team. These challenges aim to advance research across multiple themes in football, including broadcast video understanding, field understanding, and player understanding. This year, the challenges encompass four vision-based tasks. (1) Ball Action Spotting, focusing on precisely localizing when and which soccer actions related to the ball occur, (2) Dense Video Captioning, focusing on describing the broadcast with natural language and anchored timestamps, (3) Multi-View Foul Recognition, a novel task focusing on analyzing multiple viewpoints of a potential foul incident to classify whether a foul occurred and assess its severity, (4) Game State Reconstruction, another novel task focusing on reconstructing the game state from broadcast videos onto a 2D top-view map of the field. Detailed information about the tasks, challenges, and leaderboards can be found at https://www.soccer-net.org, with baselines and development kits available at https://github.com/SoccerNet., Comment: 7 pages, 1 figure

The evolution of surface layers of a glassy material heated by a laser pulse above the glass transition temperature and cooled by heat diffusion is considered as the flow of a stretchable viscous fluid. The strong dependence of viscosity on temperature and pressure leads to the appearance of a hump with reduced density. This hydrodynamic model for laser swelling is formulated in general form. We present a 1D solution for laser swelling of a thin glassy polymer film on a strongly thermally conductive substrate for laser pulses long enough that the sound confinement effect can be neglected. It is shown that within this condition the evolution of the film thickness over time can be addressed using a second-order ordinary differential equation. It is also shown that in some cases this equation can be reduced to a first-order differential equation resembling the phenomenological equation of the previously published relaxation model of laser swelling. The main features of the dependence of laser swelling on laser fluence, namely the threshold at low fluencies and saturation at high fluencies, have been clarified allowing for the dependence of viscosity on pressure, which was not taken into account in the previous theoretical studies of laser swelling. Typical regimes of the film thickness evolution are considered and compared to existing experimental data.

In these notes, we prove a recent conjecture posed in the paper by R\"ais\"a, O. et al. [Subsampling is not Magic: Why Large Batch Sizes Work for Differentially Private Stochastic Optimization (2024)]. Theorem 6.2 of the paper asserts that for the Sampled Gaussian Mechanism - a composition of subsampling and additive Gaussian noise, the effective noise level, $\sigma_{\text{eff}} = \frac{\sigma(q)}{q}$, decreases as a function of the subsampling rate $q$. Consequently, larger subsampling rates are preferred for better privacy-utility trade-offs. Our notes provide a rigorous proof of Conjecture 6.3, which was left unresolved in the original paper, thereby completing the proof of Theorem 6.2.

Speaker anonymization is an effective privacy protection solution that conceals the speaker's identity while preserving the linguistic content and paralinguistic information of the original speech. To establish a fair benchmark and facilitate comparison of speaker anonymization systems, the VoicePrivacy Challenge (VPC) was held in 2020 and 2022, with a new edition planned for 2024. In this paper, we describe our proposed speaker anonymization system for VPC 2024. Our system employs a disentangled neural codec architecture and a serial disentanglement strategy to gradually disentangle the global speaker identity and time-variant linguistic content and paralinguistic information. We introduce multiple distillation methods to disentangle linguistic content, speaker identity, and emotion. These methods include semantic distillation, supervised speaker distillation, and frame-level emotion distillation. Based on these distillations, we anonymize the original speaker identity using a weighted sum of a set of candidate speaker identities and a randomly generated speaker identity. Our system achieves the best trade-off of privacy protection and emotion preservation in VPC 2024., Comment: System description for VPC 2024

Combining high peak and high average power has long been a key challenge of ultrafast laser technology, crucial for applications such as laser-plasma acceleration and strong-field physics. A promising solution lies in post-compressed ytterbium lasers, but scaling these to high pulse energies presents a major bottleneck. Post-compression techniques, particularly Herriott-type multi-pass cells (MPCs), have enabled large peak power boosts at high average powers but their pulse energy acceptance reaches practical limits defined by setup size and coating damage threshold. In this work, we address this challenge and demonstrate a novel type of compact, energy-scalable MPC (CMPC). By employing a novel MPC configuration and folding the beam path, the CMPC introduces a new degree of freedom for downsizing the setup length, enabling compact setups even for large pulse energies. We experimentally and numerically verify the CMPC approach, demonstrating post-compression of 8 mJ pulses from 1 ps down to 51 fs in atmospheric air using a cell roughly 45 cm in length at low fluence values. Additionally, we discuss the potential for energy scaling up to 200 mJ with a setup size reaching 2.5 m. Our work presents a new approach to high-energy post-compression, with up-scaling potential far beyond the demonstrated parameters. This opens new routes for achieving the high peak and average powers necessary for demanding applications of ultrafast lasers., Comment: 18 pages, 13 figures

This paper explores the concept of external magnetic control for vine robots to enable their high curvature steering and navigation for use in endoluminal applications. Vine robots, inspired by natural growth and locomotion strategies, present unique shape adaptation capabilities that allow passive deformation around obstacles. However, without additional steering mechanisms, they lack the ability to actively select the desired direction of growth. The principles of magnetically steered growing robots are discussed, and experimental results showcase the effectiveness of the proposed magnetic actuation approach. We present a 25 mm diameter vine robot with integrated magnetic tip capsule, including 6 Degrees of Freedom (DOF) localization and camera and demonstrate a minimum bending radius of 3.85 cm with an internal pressure of 30 kPa. Furthermore, we evaluate the robot's ability to form tight curvature through complex navigation tasks, with magnetic actuation allowing for extended free-space navigation without buckling. The suspension of the magnetic tip was also validated using the 6 DOF localization system to ensure that the shear-free nature of vine robots was preserved. Additionally, by exploiting the magnetic wrench at the tip, we showcase preliminary results of vine retraction. The findings contribute to the development of controllable vine robots for endoluminal applications, providing high tip force and shear-free navigation., Comment: 13 pages, 10 figures

Classifying breast cancer molecular subtypes is crucial for tailoring treatment strategies. While immunohistochemistry (IHC) and gene expression profiling are standard methods for molecular subtyping, IHC can be subjective, and gene profiling is costly and not widely accessible in many regions. Previous approaches have highlighted the potential application of deep learning models on H&E-stained whole slide images (WSI) for molecular subtyping, but these efforts vary in their methods, datasets, and reported performance. In this work, we investigated whether H&E-stained WSIs could be solely leveraged to predict breast cancer molecular subtypes (luminal A, B, HER2-enriched, and Basal). We used 1,433 WSIs of breast cancer in a two-step pipeline: first, classifying tumor and non-tumor tiles to use only the tumor regions for molecular subtyping; and second, employing a One-vs-Rest (OvR) strategy to train four binary OvR classifiers and aggregating their results using an eXtreme Gradient Boosting (XGBoost) model. The pipeline was tested on 221 hold-out WSIs, achieving an overall macro F1 score of 0.95 for tumor detection and 0.73 for molecular subtyping. Our findings suggest that, with further validation, supervised deep learning models could serve as supportive tools for molecular subtyping in breast cancer. Our codes are made available to facilitate ongoing research and development., Comment: 16 pages, 5 figures (+4 supplementary figures), 4 tables

The estimation of 3D human poses from images has progressed tremendously over the last few years as measured on standard benchmarks. However, performance in the open world remains underexplored, as current benchmarks cannot capture its full extent. Especially in safety-critical systems, it is crucial that 3D pose estimators are audited before deployment, and their sensitivity towards single factors or attributes occurring in the operational domain is thoroughly examined. Nevertheless, we currently lack a benchmark that would enable such fine-grained analysis. We thus present STAGE, a GenAI data toolkit for auditing 3D human pose estimators. We enable a text-to-image model to control the 3D human body pose in the generated image. This allows us to create customized annotated data covering a wide range of open-world attributes. We leverage STAGE and generate a series of benchmarks to audit the sensitivity of popular pose estimators towards attributes such as gender, ethnicity, age, clothing, location, and weather. Our results show that the presence of such naturally occurring attributes can cause severe degradation in the performance of pose estimators and leads us to question if they are ready for open-world deployment.

The COMET metric has blazed a trail in the machine translation community, given its strong correlation with human judgements of translation quality. Its success stems from being a modified pre-trained multilingual model finetuned for quality assessment. However, it being a machine learning model also gives rise to a new set of pitfalls that may not be widely known. We investigate these unexpected behaviours from three aspects: 1) technical: obsolete software versions and compute precision; 2) data: empty content, language mismatch, and translationese at test time as well as distribution and domain biases in training; 3) usage and reporting: multi-reference support and model referencing in the literature. All of these problems imply that COMET scores are not comparable between papers or even technical setups and we put forward our perspective on fixing each issue. Furthermore, we release the sacreCOMET package that can generate a signature for the software and model configuration as well as an appropriate citation. The goal of this work is to help the community make more sound use of the COMET metric.

We present the discovery of 118 new ultracool dwarf candidates, discovered using a new machine learning tool, named \texttt{SMDET}, applied to time series images from the Wide-field Infrared Survey Explorer. We gathered photometric and astrometric data to estimate each candidate's spectral type, distance, and tangential velocity. This sample has a photometrically estimated spectral class distribution of 28 M dwarfs, 64 L dwarfs, and 18 T dwarfs. We also identify a T subdwarf candidate, two extreme T subdwarf candidates, and two candidate young ultracool dwarfs. Five objects did not have enough photometric data for any estimations to be made. To validate our estimated spectral types, spectra were collected for 2 objects, yielding confirmed spectral types of T5 (estimated T5) and T3 (estimated T4). Demonstrating the effectiveness of machine learning tools as a new large-scale discovery technique., Comment: 14 pages, 8 figures, 2 tables, extended table 1, accepted to Astronomical Journal

This paper presents a performance comparison of various MIMO radar multiplexing approaches where the increasing number of transmitters adversely affects the dynamic range of the resultant MIMO system. The investigated multiplexing techniques are code-division multiplexing in phase-coded frequency-modulated continuous wave (FMCW) and phase-modulated continuous wave (PMCW) radar. Additionally, random fast/slow time multiplexing in time and frequency is considered for frequency-modulated continuous wave and orthogonal frequency-division multiplexing (OFDM) radars. The comparative analysis is conducted through simulations, evaluating the scalability with the number of transmitters and addressing other pertinent implementation concerns. The findings provide insights into the trade-offs associated with each approach, aiding in selecting suitable MIMO radar multiplexing strategies for practical applications., Comment: International Conference on Microwaves for Intelligent Mobility (ICMIM 2024)

We suggest a novel algorithm for online change point detection based on sequential score function estimation and tracking the best expert approach. The core of the procedure is a version of the fixed share forecaster for the case of infinite number of experts and quadratic loss functions. The algorithm shows a promising performance in numerical experiments on artificial and real-world data sets. We also derive new upper bounds on the dynamic regret of the fixed share forecaster with varying parameter, which are of independent interest., Comment: 43 pages, 4 figures

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

In this note we construct a formulation of 4d pure Yang-Mills theory in the unfolded language of higher-spin gravity. It represents a first-order formulation possessing manifest diffeomorphism- and gauge-invariance. We make use of an unfolding method that was recently put forward and used to unfold scalar electrodynamics. We also discuss features of various unfolding maps defined by the unfolded equations., Comment: 12 pages

Existing techniques for monocular 3D detection have a serious restriction. They tend to perform well only on a limited set of benchmarks, faring well either on ego-centric car views or on traffic camera views, but rarely on both. To encourage progress, this work advocates for an extended evaluation of 3D detection frameworks across different camera perspectives. We make two key contributions. First, we introduce the CARLA Drone dataset, CDrone. Simulating drone views, it substantially expands the diversity of camera perspectives in existing benchmarks. Despite its synthetic nature, CDrone represents a real-world challenge. To show this, we confirm that previous techniques struggle to perform well both on CDrone and a real-world 3D drone dataset. Second, we develop an effective data augmentation pipeline called GroundMix. Its distinguishing element is the use of the ground for creating 3D-consistent augmentation of a training image. GroundMix significantly boosts the detection accuracy of a lightweight one-stage detector. In our expanded evaluation, we achieve the average precision on par with or substantially higher than the previous state of the art across all tested datasets.

This paper provides a detailed examination of the advancements and applications of large language models in the healthcare sector, with a particular emphasis on clinical applications. The study traces the evolution of LLMs from their foundational technologies to the latest developments in domain-specific models and multimodal integration. It explores the technical progression from encoder-based models requiring fine-tuning to sophisticated approaches that integrate textual, visual, and auditory data, thereby facilitating comprehensive AI solutions in healthcare. The paper discusses both the opportunities these technologies present for enhancing clinical efficiency and the challenges they pose in terms of ethics, data privacy, and implementation. Additionally, it critically evaluates the deployment strategies of LLMs, emphasizing the necessity of open-source models to ensure data privacy and adaptability within healthcare environments. Future research directions are proposed, focusing on empirical studies to evaluate the real-world efficacy of LLMs in healthcare and the development of open datasets for further research. This review aims to provide a comprehensive resource for both newcomers and multidisciplinary researchers interested in the intersection of AI and healthcare., Comment: Submitted to PLOS Digital Health

We present a novel multiferroic double perovskite compound, CaZnFeOsO$_6$ (CZFOO), exhibiting combined ferroelectric and ferrimagnetic properties. Through ab initio density functional theory calculations, we predict CZFOO as a unique example of an A-site and B-site ordered double perovskite structure, AA'BB'O$_6$. In this compound, Fe$^{3+}$ and Os$^{5+}$ ions generate substantial magnetization, while Ca$^{2+}$ and Zn$^{2+}$ ions create a layerwise polar environment, resulting in a synergistic combination for multiferroicity. We determine the magnitude of the spontaneous polarization, $\vert P_s \vert$, to be 16.8 $\mu$ C/cm$^2$, and the magnetic moment is approximately 2 $\mu_B$ per formula unit. The remarkable ferroelectric and ferrimagnetic behaviors exhibited by CZFOO make it a promising candidate for various device applications. Despite the significant magnetization and polarization observed, surpassing those of other double perovskites, we find a weak spin-orbit coupling, leading to the absence of any significant magnetoelectric effect in CZFOO. Our findings shed light on the potential of CZFOO as a multiferroic material and provide insights into the intricate interplay between ferroelectricity and ferrimagnetism in double perovskite compounds.

We show a Springer type theorem for the variety of parabolic subgroups of type $1,2,6$ for all groups of type $E_6$. As far as we know this gives the first example for the validity of the Springer theorem for projective homogeneous varieties of type $^2E_6$ different from varieties of Borel subgroups. The proof combines several topics, notably the Rost invariant, a Tits construction, Cartan's symmetric spaces and indirectly the structure of the Chow motives of projective homogeneous varieties of exceptional type., Comment: 8 pages

Wireless power transfer is one of key drivers in modern consumer electronics, as it allows enhancing the convenience and usability of many devices. However, in most cases wireless charging is accessible only to devices developed with such a functionality in mind or at least to gadgets with standard charging connectors, such as USB type-C. We propose the design of a rechargeable battery having the size and output voltage characteristics of a standard AA battery, but supporting magnetic wireless power transfer from charging stations of widely used Qi standard. In our numerical simulations, we study the magnetic field distributions and scattering parameters of the introduced battery with a curved receiving coil coupled to a planar transmitting coil. The obtained results are supported by an experimental realization of the proposed design and a study of the prototype characteristics. The developed battery design facilitates the possibility of adding wireless charging functionality to a wide range of electronic devices in a universal way., Comment: 8 pages, 5 figures + Supplementary Materials

Hydrogen-deficient Carbon (HdC) stars are rare, low-mass, chemically peculiar, supergiant variables believed to be formed by a double white dwarf (DWD) merger, specifically of a Carbon/Oxygen- (CO-) and a Helium-white dwarf (He-WD). They consist of two subclasses -- the dust-producing R Coronae Borealis (RCB) variables and their dustless counterparts the dustless HdCs (dLHdCs). Additionally, there is another, slightly cooler set of potentially related carbon stars, the DY Persei type variables which have some, but not conclusive, evidence of Hydrogen-deficiency. Recent works have begun to explore the relationship between these three classes of stars, theorizing that they share an evolutionary pathway (a DWD merger) but come from different binary populations, specifically different total masses (M$_{\rm tot}$) and mass ratios ($q$). In this work, we use the MESA modelling framework that has previously been used to model RCB stars and vary the merger parameters, M$_{\rm tot}$ and $q$, to explore how those parameters affect the abundances, temperatures, and luminosities of the resultant post-merger stars. We find that lower M$_{\rm tot}$ and larger $q$'s both decrease the luminosity and temperatures of post-merger models to the region of the Hertzsprung-Russell Diagram populated by the dLHdCs. These lower M$_{\rm tot}$ and larger $q$ models also have smaller oxygen isotopic ratios ($^{16}$O/$^{18}$O) which is consistent with recent observations of dLHdCs compared to RCBs. None of the models generated in this work can explain the existence of the DY Persei type variables, however this may arise from the assumed metallicity of the models., Comment: 10 pages, 6 figures, re-submitted to MNRAS, has undergone one round of positive review

We compute the Chow ring of a quasi-split geometrically almost simple algebraic group assuming the coefficients to be a field. This extends the classical computation for split groups done by Kac to the non-split quasi-split case. For the proof we introduce and study equivariant conormed Chow rings, which are well adapted to the study of quasi-split groups and their homogeneous varieties., Comment: 62 pages, v2: minor revision

As NLP systems are increasingly deployed at scale, concerns about their potential negative impacts have attracted the attention of the research community, yet discussions of risk have mostly been at an abstract level and focused on generic AI or NLP applications. We argue that clearer assessments of risks and harms to users--and concrete strategies to mitigate them--will be possible when we specialize the analysis to more concrete applications and their plausible users. As an illustration, this paper is grounded in cooking recipe procedural document question answering (ProcDocQA), where there are well-defined risks to users such as injuries or allergic reactions. Our case study shows that an existing language model, applied in "zero-shot" mode, quantitatively answers real-world questions about recipes as well or better than the humans who have answered the questions on the web. Using a novel questionnaire informed by theoretical work on AI risk, we conduct a risk-oriented error analysis that could then inform the design of a future system to be deployed with lower risk of harm and better performance.

Teams can outperform individuals; could adding AI teammates further bolster performance of teams solving problems collaboratively? Collaborative problem solving (CPS) research commonly studies teams with two agents (human-human or human-AI), but team research literature finds that, for complex tasks, larger teams are more effective. Progress in studying collaboration with more than two agents, through textual records of team interactions, is hindered by a major data challenge: available CPS corpora are predominantly dyadic, and adapting pre-existing CPS tasks to more agents is non-trivial. We address this data challenge by developing a CPS task generator, CPS-TaskForge, that can produce environments for studying CPS under a wide array of conditions, and releasing a CPS task design checklist grounded in the theoretical PISA 2015 CPS framework to help facilitate the development of CPS corpora with more agents. CPS-TaskForge takes the form of a resource management (tower defense) game, and different CPS tasks can be studied by manipulating game design parameters. We conduct a case study with groups of 3-4 humans to validate production of diverse natural language CPS communication in a game instance produced by CPS-TaskForge. We discuss opportunities for advancing research in CPS (both with human-only and human-AI teams) using different task configurations. We will release data and code.

The use of decision trees for percolation inequalities started with the celebrated O'Donnell--Saks--Schramm--Servedio (OSSS) inequality. We prove decision tree generalizations of the Harris--Kleitman (HK), van den Berg--Kesten (vdBK), and other inequalities. These inequalities are then applied to estimate the connection probabilities in Bernoulli bond percolation on general graphs., Comment: 20 pages

As the complexities of urban environments continue to grow, the modelling of transportation systems become increasingly challenging. This paper explores the application of advanced Graph Neural Network (GNN) architectures as surrogate models for strategic transport planning. Building upon a prior work that laid the foundation with graph convolution networks (GCN), our study delves into the comparative analysis of established GCN with the more expressive Graph Attention Network (GAT). Additionally, we propose a novel GAT variant (namely GATv3) to address over-smoothing issues in graph-based models. Our investigation also includes the exploration of a hybrid model combining both GCN and GAT architectures, aiming to investigate the performance of the mixture. The three models are applied to various experiments to understand their limits. We analyse hierarchical regression setups, combining classification and regression tasks, and introduce fine-grained classification with a proposal of a method to convert outputs to precise values. Results reveal the superior performance of the new GAT in classification tasks. To the best of the authors' knowledge, this is the first GAT model in literature to achieve larger depths. Surprisingly, the fine-grained classification task demonstrates the GCN's unexpected dominance with additional training data. This shows that synthetic data generators can increase the training data, without overfitting issues whilst improving model performance. In conclusion, this research advances GNN based surrogate modelling, providing insights for refining GNN architectures. The findings open avenues for investigating the potential of the newly proposed GAT architecture and the modelling setups for other transportation problems.