Search

Synthesizing a novel view from a single input image is a challenging task. Traditionally, this task was approached by estimating scene depth, warping, and inpainting, with machine learning models enabling parts of the pipeline. More recently, generative models are being increasingly employed in novel view synthesis (NVS), often encompassing the entire end-to-end system. In this work, we adapt a modern diffusion model architecture for end-to-end NVS in the pixel space, substantially outperforming previous state-of-the-art (SOTA) techniques. We explore different ways to encode geometric information into the network. Our experiments show that while these methods may enhance performance, their impact is minor compared to utilizing improved generative models. Moreover, we introduce a novel NVS training scheme that utilizes single-view datasets, capitalizing on their relative abundance compared to their multi-view counterparts. This leads to improved generalization capabilities to scenes with out-of-domain content.

This paper presents a novel approach to building mission planners based on neural networks with Transformer architecture and Large Language Models (LLMs). This approach demonstrates the possibility of setting a task for a mobile robot and its successful execution without the use of perception algorithms, based only on the data coming from the camera. In this work, a success rate of more than 50\% was obtained for one of the basic actions for mobile robots. The proposed approach is of practical importance in the field of warehouse logistics robots, as in the future it may allow to eliminate the use of markings, LiDARs, beacons and other tools for robot orientation in space. In conclusion, this approach can be scaled for any type of robot and for any number of robots.

Wire bending is a technique used in manufacturing to mass-produce items such as clips, mounts, and braces. Wire bending machines like the DIWire by Pensalabs have made this process accessible for personal fabrication. However, such machines are controlled using Computer Aided Manufacturing (CAM) software which is hard to use, making custom design challenging. We present Y-AR, a Computer Aided Design (CAD) interface for 3D wire bending. Y-AR uses mixed reality to let designers create structures that physically connect to objects in the environment. The interface incorporates springs as design primitives which (1) apply forces to hold objects, and (2) counter-act dimensional inaccuracies inherently caused by mid air modeling and measurement errors in AR. We demonstrate design workflows to design and fabricate a range of mechanisms designed in Y-AR as well as structures made using free-hand design tools. In our usability evaluation, all 12 participants successfully designed and fabricated a functional bottle holder with Y-AR. 10 out of 12 participants felt that the system aided their design process, rating it above 7 out of 10.

To evaluate the societal impacts of GenAI requires a model of how social harms emerge from interactions between GenAI, people, and societal structures. Yet a model is rarely explicitly defined in societal impact evaluations, or in the taxonomies of societal impacts that support them. In this provocation, we argue that societal impacts should be conceptualised as application- and context-specific, incommensurable, and shaped by questions of social power. Doing so leads us to conclude that societal impact evaluations using existing taxonomies are inherently limited, in terms of their potential to reveal how GenAI systems may interact with people when introduced into specific social contexts. We therefore propose a governance-first approach to managing societal harms attended by GenAI technologies., Comment: 3 pages

Lately, there has been a surge in interest surrounding generative modeling of time series data. Most existing approaches are designed either to process short sequences or to handle long-range sequences. This dichotomy can be attributed to gradient issues with recurrent networks, computational costs associated with transformers, and limited expressiveness of state space models. Towards a unified generative model for varying-length time series, we propose in this work to transform sequences into images. By employing invertible transforms such as the delay embedding and the short-time Fourier transform, we unlock three main advantages: i) We can exploit advanced diffusion vision models; ii) We can remarkably process short- and long-range inputs within the same framework; and iii) We can harness recent and established tools proposed in the time series to image literature. We validate the effectiveness of our method through a comprehensive evaluation across multiple tasks, including unconditional generation, interpolation, and extrapolation. We show that our approach achieves consistently state-of-the-art results against strong baselines. In the unconditional generation tasks, we show remarkable mean improvements of 58.17% over previous diffusion models in the short discriminative score and 132.61% in the (ultra-)long classification scores. Code is at https://github.com/azencot-group/ImagenTime., Comment: Accepted to NeurIPS 2024; The first two authors contributed equally

This study explores the ability of Large Language Model (LLM) agents to detect and correct hallucinations in AI-generated content. A primary agent was tasked with creating a blog about a fictional Danish artist named Flipfloppidy, which was then reviewed by another agent for factual inaccuracies. Most LLMs hallucinated the existence of this artist. Across 4,900 test runs involving various combinations of primary and reviewing agents, advanced AI models such as Llama3-70b and GPT-4 variants demonstrated near-perfect accuracy in identifying hallucinations and successfully revised outputs in 85% to 100% of cases following feedback. These findings underscore the potential of advanced AI models to significantly enhance the accuracy and reliability of generated content, providing a promising approach to improving AI workflow orchestration.

The aim of this work is to learn models of population dynamics of physical systems that feature stochastic and mean-field effects and that depend on physics parameters. The learned models can act as surrogates of classical numerical models to efficiently predict the system behavior over the physics parameters. Building on the Benamou-Brenier formula from optimal transport and action matching, we use a variational problem to infer parameter- and time-dependent gradient fields that represent approximations of the population dynamics. The inferred gradient fields can then be used to rapidly generate sample trajectories that mimic the dynamics of the physical system on a population level over varying physics parameters. We show that combining Monte Carlo sampling with higher-order quadrature rules is critical for accurately estimating the training objective from sample data and for stabilizing the training process. We demonstrate on Vlasov-Poisson instabilities as well as on high-dimensional particle and chaotic systems that our approach accurately predicts population dynamics over a wide range of parameters and outperforms state-of-the-art diffusion-based and flow-based modeling that simply condition on time and physics parameters.

Julia has been heralded as a potential successor to Python for scientific machine learning and numerical computing, boasting ergonomic and performance improvements. Since Julia's inception in 2012 and declaration of language goals in 2017, its ecosystem and language-level features have grown tremendously. In this paper, we take a modern look at Julia's features and ecosystem, assess the current state of the language, and discuss its viability and pitfalls as a replacement for Python as the de-facto scientific machine learning language. We call for the community to address Julia's language-level issues that are preventing further adoption., Comment: To appear at the 2024 NeurIPS Machine Learning and the Physical Sciences Workshop

We derive two rigorous constraints on the spectrum of massive states in weakly coupled theories with massless scalars in the adjoint representation of a large-$N$ gauge group. First, we show that the presence of massive spinning states necessitates the existence of lighter states with lower spins. Explicitly, if there exists a massive state with spin $J > 2$, then there must be a state with spin $J-1$ and a non-zero mass lower than that of the lightest spin-$J$ state, a state with spin $J-2$ and a mass lower than that of the lightest spin-($J-1$) particle and so on until we reach a mass below which only states with spin less than 2 are exchanged. Second, we find strict upper bounds on the masses of the lightest states at any spin. If there are spin-$J$ states in the spectrum, the maximum mass of the lightest spin-($J+1$) state is determined by the masses of the lightest spin-$J$ and $(J-1)$ states. In the approximation that this bound applies to pion scattering in real world QCD, we find it gives a window of only ${\sim}150$ MeV for the expected mass of the yet unmeasured spin-7 meson., Comment: 26+5 pages, 2 figures

We study the geometry of the Gromoll-Meyer sphere, one of Milnor's exotic $7$-spheres. We focus on a Kaluza-Klein Ansatz, with a round $S^4$ as base space, unit $S^3$ as fibre, and $k=1,2$ $SU(2)$ instantons as gauge fields, where all quantities admit an elegant description in quaternionic language. The metric's moduli space coincides with the $k=1,2$ instantons' moduli space quotiented by the isometry of the base, plus an additional $\mathbb{R}^+$ factor corresponding to the radius of the base, $r$. We identify a "center" of the $k=2$ instanton moduli space with enhanced symmetry. This $k=2$ solution is used together with the maximally symmetric $k=1$ solution to obtain a metric of maximal isometry, $SO(3)\times O(2)$, and to compute its Ricci tensor explicitly. This allows us to put a bound on $r$ to ensure positive Ricci curvature, which implies various energy conditions for an $8$-dimensional static space-time. The construction opens for a concrete examination of the properties of the sectional curvature., Comment: 33 pages, 4 figures

We address the problem of indirect quantization of a source subject to a mean-squared error distortion constraint. A well-known result of Wolf and Ziv is that the problem can be reduced to a standard (direct) quantization problem via a two-step approach: first apply the conditional expectation estimator, obtaining a ``new'' source, then solve for the optimal quantizer for the latter source. When quantization is implemented in hardware, however, invariably constraints on the allowable class of quantizers are imposed, typically limiting the class to \emph{time-invariant} scalar quantizers with contiguous quantization cells. In the present work, optimal indirect quantization subject to these constraints is considered. Necessary conditions an optimal quantizer within this class must satisfy are derived, in the form of generalized Lloyd-Max conditions, and an iterative algorithm for the design of such quantizers is proposed. Furthermore, for the case of a scalar observation, we derive a non-iterative algorithm for finding the optimal indirect quantizer based on dynamic programming.

We predict Bose-Einstein condensation and superfluidity of dipolar excitons, formed by electron-hole pairs in spatially separated gapped hexagonal $\alpha-T_{3}$ (GHAT3) layers. In the $\alpha-T_{3}$ model, the AB-honeycomb lattice structure is supplemented with C atoms located at the centers of the hexagons in the lattice. We considered the $\alpha-T_{3}$ model in the presence of a mass term which opens a gap in the energy dispersive spectrum. The gap opening mass term, caused by a weak magnetic field, plays the role of Zeeman splitting at low magnetic fields for this pseudospin-1 system. The band structure of GHAT3 monolayers leads to the formation of two distinct types of excitons in the GHAT3 double layer. We consider two types of dipolar excitons in double-layer GHAT3: (a) ``A excitons'', which are bound states of electrons in the conduction band (CB) and holes in the intermediate band (IB) and (b) ``B excitons'', which are bound states of electrons in the CB and holes in the valence band (VB). The binding energy of A and B dipolar excitons is calculated. For a two-component weakly interacting Bose gas of dipolar excitons in a GHAT3 double layer, we obtain the energy dispersion of collective excitations, the sound velocity, the superfluid density, and the mean-field critical temperature $T_{c}$ for superfluidity., Comment: 17 pages, 14 figures

Connected and automated vehicles and robot swarms hold transformative potential for enhancing safety, efficiency, and sustainability in the transportation and manufacturing sectors. Extensive testing and validation of these technologies is crucial for their deployment in the real world. While simulations are essential for initial testing, they often have limitations in capturing the complex dynamics of real-world interactions. This limitation underscores the importance of small-scale testbeds. These testbeds provide a realistic, cost-effective, and controlled environment for testing and validating algorithms, acting as an essential intermediary between simulation and full-scale experiments. This work serves to facilitate researchers' efforts in identifying existing small-scale testbeds suitable for their experiments and provide insights for those who want to build their own. In addition, it delivers a comprehensive survey of the current landscape of these testbeds. We derive 62 characteristics of testbeds based on the well-known sense-plan-act paradigm and offer an online table comparing 22 small-scale testbeds based on these characteristics. The online table is hosted on our designated public webpage www.cpm-remote.de/testbeds, and we invite testbed creators and developers to contribute to it. We closely examine nine testbeds in this paper, demonstrating how the derived characteristics can be used to present testbeds. Furthermore, we discuss three ongoing challenges concerning small-scale testbeds that we identified, i.e., small-scale to full-scale transition, sustainability, and power and resource management., Comment: 16 pages, 11 figures, 1 table. This work has been submitted to the IEEE Robotics & Automation Magazine for possible publication

We present the properties of two candidate massive ($M_\star\sim10^{11}M_\odot$) and dusty ($A_{\rm v}>2.5$ mag) galaxies at $z=5-7$ in the first 0.28 deg$^2$ of the COSMOS-Web survey. One object is spectroscopically confirmed at $z_{\rm spec}=5.051$, while the other has a robust $z_{\rm phot}=6.7\pm0.3$. Thanks to their extremely red colors ($F277W-F444W\sim1.7$ mag), these galaxies satisfy the nominal color-selection for the widely-studied ``little red dot" (LRD) population with the exception of their spatially-resolved morphologies. The morphology of our targets allows us to conclude that their red continuum is dominated by highly obscured stellar emission and not by reddened nuclear activity. Using a variety of SED-fitting tools and star formation histories, we estimate the stellar masses to be $\log(M_\star)=11.32^{+0.07}_{-0.15}$ $M_\odot$ and $\log(M_\star)=11.2^{+0.1}_{-0.2}$ $M_\odot$, respectively, with a red continuum emission dominated by a recent episode of star formation. We then compare their number density to the halo mass function to infer stellar baryon fractions of $\epsilon_\star\sim0.25$ and $\epsilon_\star\sim0.5$. Both are significantly higher than what is commonly observed in lower-z galaxies or more dust-obscured galaxies at similar redshifts. With very bright ultra-high-z Lyman-Break Galaxies and some non-AGN dominated LRDs, such ``extended" LRDs represent another population that may require very efficient star formation at early times., Comment: 11 pages, 4 figures, 1 table. Accepted for publication in ApJL

Recent ethnographic research reveals that gang dynamics in Chicago's Southside have evolved with decentralized micro-gang "set" factions and cross-gang interpersonal networks marking the contemporary landscape. However, standard police datasets lack the depth to analyze gang violence with such granularity. To address this, we employed a natural language processing strategy to analyze text from a Chicago gangs message board. By identifying proper nouns, probabilistically linking them to gang sets, and assuming social connections among names mentioned together, we created a social network dataset of 271 individuals across 11 gang sets. Using Louvain community detection, we found that these individuals often connect with gang-affiliated peers from various gang sets that are physically proximal. Hierarchical logistic regression revealed that individuals with ties to homicide victims and central positions in the overall gang network were at increased risk of victimization, regardless of gang affiliation. This research demonstrates that utilizing crowd-sourced information online can enable the study of otherwise inaccessible topics and populations., Comment: 24 pages, 5 figures

Temporal conceptual data modelling, as an extension to regular conceptual data modelling languages such as EER and UML class diagrams, has received intermittent attention across the decades. It is receiving renewed interest in the context of, among others, business process modelling that needs robust expressive data models to complement them. None of the proposed temporal conceptual data modelling languages have been tested on understandability and usability by modellers, however, nor is it clear which temporal constraints would be used by modellers or whether the ones included are the relevant temporal constraints. We therefore sought to investigate temporal representations in temporal conceptual data modelling languages, design a, to date, most expressive language, TREND, through small-scale qualitative experiments, and finalise the graphical notation and modelling and understanding in large scale experiments. This involved a series of 11 experiments with over a thousand participants in total, having created 246 temporal conceptual data models. Key outcomes are that choice of label for transition constraints had limited impact, as did extending explanations of the modelling language, but expressing what needs to be modelled in controlled natural language did improve model quality. The experiments also indicate that more training may be needed, in particular guidance for domain experts, to achieve adoption of temporal conceptual data modelling by the community., Comment: 44 pages, 23 figures, 11 tables, submitted to an international journal

We study the geometric structure of Poncelet $n$-gons from a projective point of view. In particular we present explicit constructions of Poncelet $n$-gons for certain $n$ and derive algebraic characterisations in terms of bracket polynomials. Via the connections of Poncelet polygons and $(N_4)$-configurations, the results of this article can be used to construct a large class of specific movable $(N_4)$-configurations, the trivial celestial 4-configurations, which up to this point were all thought to be rigid and to require regular polygons for their construction., Comment: 35 pages, 20 figures

We study relations between $(n_4)$ incidence configurations and the classical Poncelet Porism. Poncelet's result studies two conics and a sequence of points and lines that inscribes one conic and circumscribes the other. Poncelet's Porism states that whether this sequence closes up after $m$ steps only depends on the conics and not on the initial point of the sequence. In other words: Poncelet polygons are movable. We transfer this motion into a flexibility statement about a large class of $(n_4)$ configurations, which are configurations where 4 (straight) lines pass through each point and four points lie on each line. A first instance of such configurations in real geometry had been given by Gr\"unbaum and Rigby in their classical 1990 paper where they constructed the first known real geometric realisation of a well known combinatorial $(21_4)$ configuration (which had been studied by Felix Klein), now called the Gr\"unbaum-Rigby configuration. Since then, there has been an intensive search for movable $(n_4)$ configurations, but it is very surprising that the Gr\"unbaum-Rigby $(21_4)$ configuration admits nontrivial motions. It is well-known that the Gr\"unbaum-Rigby configuration is the smallest example of an infinite class of $(n_4)$ configurations, the trivial celestial configurations. A major result of this paper is that we show that all trivial celestial configurations are movable via Poncelet's Porism and results about properties of Poncelet grids. Alternative approaches via geometry of billiards, in-circle nets, and pentagram maps that relate the subject to discrete integrable systems are given as well., Comment: 56 pages, 34 figures

Quantum computers must achieve large-scale, fault-tolerant operation to deliver on their promise of transformational processing power [1-4]. This will require thousands or millions of high-fidelity quantum gates and similar numbers of qubits [5]. Demonstrations using neutral-atom qubits trapped and manipulated by lasers have shown that this modality can provide high two-qubit gate (CZ) fidelities and scalable operation [6-10]. However, the gates in these demonstrations are driven by lasers that do not resolve individual qubits, with universal computation enabled by physical mid-circuit shuttling of the qubits. This relatively slow operation will greatly extend runtimes for useful, large-scale computation. Here we demonstrate a universal neutral-atom quantum computer with gate rates limited by optical switching times, rather than shuttling, by individually addressing tightly focused laser beams at an array of single atoms. We achieve CZ fidelity of 99.35(4)% and local single qubit RZ gate fidelity of 99.902(8)%. Moreover, we demonstrate non-destructive readout of alkali-atom qubits with sub-percent loss, which boosts operational speed. This technique also enables us to measure 99.73(3)% CZ fidelity with atom-loss events excluded, which is a record among long lived neutral-atom qubits and highlights the path to higher fidelity and error correction. Our results represent a critical step towards large-scale, fault-tolerant neutral-atom quantum computers that can execute computations on practical timescales.

In a paper from 2015, Ding et al. (IEEE Trans. IT, May 2015) conjectured that for odd $m$, the minimum distance of the binary BCH code of length $2^m-1$ and designed distance $2^{m-2}+1$ is equal to the Bose distance calculated in the same paper. In this paper, we prove the conjecture. In fact, we prove a stronger result: the weight of the generating idempotent is equal to the Bose distance for both odd and even $m$. Our main tools are some new properties of the so-called fibbinary integers, in particular, the splitting field of related polynomials, and the relation of these polynomials to the idempotent of the BCH code., Comment: Some small simplifications. Also, added concluding remarks

Deep learning is mainly based on utilizing gradient-based optimization for training Deep Neural Network (DNN) models. Although robust and widely used, gradient-based optimization algorithms are prone to getting stuck in local minima. In this modern deep learning era, the state-of-the-art DNN models have millions and billions of parameters, including weights and biases, making them huge-scale optimization problems in terms of search space. Tuning a huge number of parameters is a challenging task that causes vanishing/exploding gradients and overfitting; likewise, utilized loss functions do not exactly represent our targeted performance metrics. A practical solution to exploring large and complex solution space is meta-heuristic algorithms. Since DNNs exceed thousands and millions of parameters, even robust meta-heuristic algorithms, such as Differential Evolution, struggle to efficiently explore and converge in such huge-dimensional search spaces, leading to very slow convergence and high memory demand. To tackle the mentioned curse of dimensionality, the concept of blocking was recently proposed as a technique that reduces the search space dimensions by grouping them into blocks. In this study, we aim to introduce Histogram-based Blocking Differential Evolution (HBDE), a novel approach that hybridizes gradient-based and gradient-free algorithms to optimize parameters. Experimental results demonstrated that the HBDE could reduce the parameters in the ResNet-18 model from 11M to 3K during the training/optimizing phase by metaheuristics, namely, the proposed HBDE, which outperforms baseline gradient-based and parent gradient-free DE algorithms evaluated on CIFAR-10 and CIFAR-100 datasets showcasing its effectiveness with reduced computational demands for the very first time., Comment: 8 pages, 5 figures, 3 tables, accepted at IEEE CCECE 2024 (updated Fig. 1 and conclusion remarks)

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Earthquake occurrence is notoriously difficult to predict. While some aspects of their spatiotemporal statistics can be relatively well captured by point-process models, very little is known regarding the magnitude of future events, and it is deeply debated whether it is possible to predict the magnitude of an earthquake before it starts. This is due both to the lack of information about fault conditions and to the inherent complexity of rupture dynamics. Consequently, even state of the art forecasting models typically assume no knowledge about the magnitude of future events besides the time-independent Gutenberg Richter (GR) distribution, which describes the marginal distribution over large regions and long times. This approach implicitly assumes that earthquake magnitudes are independent of previous seismicity and are identically distributed. In this work we challenge this view by showing that information about the magnitude of an upcoming earthquake can be directly extracted from the seismic history. We present MAGNET - MAGnitude Neural EsTimation model, an open-source, geophysically-inspired neural-network model for probabilistic forecasting of future magnitudes from cataloged properties: hypocenter locations, occurrence times and magnitudes of past earthquakes. Our history-dependent model outperforms stationary and quasi-stationary state of the art GR-based benchmarks, in real catalogs in Southern California, Japan and New-Zealand. This demonstrates that earthquake catalogs contain information about the magnitude of future earthquakes, prior to their occurrence. We conclude by proposing methods to apply the model in characterization of the preparatory phase of earthquakes, and in operational hazard alert and earthquake forecasting systems., Comment: 4 main figure, 1 main table

In this paper, we establish a general setup for studying incidence-preserving motions of projective geometric configurations of points and lines via a "projective rigidity matrix". The spaces of infinitesimal motions of a point-line configuration and dependencies amongst the point-line incidences can be interpreted as the kernel and co-kernel of this projective rigidity matrix, respectively. We also introduce a symmetry-adapted projective rigidity matrix for analysing symmetric configurations and their symmetry-preserving motions. The symmetry may be a point group or a more general symmetry, such as an autopolarity.

Prior research has enhanced the ability of Large Language Models (LLMs) to solve logic puzzles using techniques such as chain-of-thought prompting or introducing a symbolic representation. These frameworks are still usually insufficient to solve complicated logical problems, such as Zebra puzzles, due to the inherent complexity of translating natural language clues into logical statements. We introduce a multi-agent system, ZPS, that integrates LLMs with an off the shelf theorem prover. This system tackles the complex puzzle-solving task by breaking down the problem into smaller, manageable parts, generating SMT (Satisfiability Modulo Theories) code to solve them with a theorem prover, and using feedback between the agents to repeatedly improve their answers. We also introduce an automated grid puzzle grader to assess the correctness of our puzzle solutions and show that the automated grader is reliable by evaluating it in a user-study. Our approach shows improvement in all three LLMs we tested, with GPT-4 showing 166% improvement in the number of fully correct solutions.

The paper describes a number of dialogue phenomena associated with negotiative dialogue, as implemented in a development version of the Talkamatic Dialogue Manager (TDM). This implementation is an initial step towards full coverage of general features of negotiative dialogue in TDM.

A/B testers conducting large-scale tests prioritize lifts and want to be able to control false rejections of the null. This work develops a decision-theoretic framework for maximizing profits subject to false discovery rate (FDR) control. We build an empirical Bayes solution for the problem via the greedy knapsack approach. We derive an oracle rule based on ranking the ratio of expected lifts and the cost of wrong rejections using the local false discovery rate (lfdr) statistic. Our oracle decision rule is valid and optimal for large-scale tests. Further, we establish asymptotic validity for the data-driven procedure and demonstrate finite-sample validity in experimental studies. We also demonstrate the merit of the proposed method over other FDR control methods. Finally, we discuss an application to actual Optimizely experiments.

Two players play a game by alternately splitting a surface of a compact $2$-manifold along a simple closed curve that is not null-homotopic and attaching disks to the resulting boundary; the last player who can move wins. Starting from an orientable surface, the $G$-series is $01\dot{2}\dot{0}$ according to increasing genus. Starting from a nonorientable surface, the $G$-series is $012\dot{4}60\dot{3}$ according to increasing genus. Nim addition determines the $G$-values of the remaining compact $2$-manifolds., Comment: 12 pages. To be submitted to the Bulletin of The Institute for Combinatorics and its Applications This version corrects the email: lleonardjr@icloud.com

Let (X,D) be a projective log pair over the ring of integers of a number field such that the log canonical line bundle K_(X,D) or its dual -K_(X,D) is relatively ample. We introduce a canonical height of K_(X,D) (and -K(X,D)) which is finite precisely when the complexifications of K_(X,D) (and -K(X,D)) are K-semistable. When the complexifications are K-polystable, the canonical height is the height of K_(X,D) (and -K(X,D)) wrt any volume-normalized K\"ahler-Einstein metric on the complexifications of K_(X,D) (and -K(X,D)) The canonical height is shown to have a number of useful variational properties. Moreover, it may be expressed as a limit of periods on the N-fold products of the complexifications of X, as N tends to infinity. In particular, using this limit formula, the canonical height for the arithmetic log surfaces (P_1,D) over the integers, where D has at most three components, is computed explicitly in terms of the Hurwitz zeta function and its derivative at s=-1. Combining this explicit formula with previous height formulas for quaternionic Shimura curves yields a procedure for extracting information about the canonical integral models of some Shimura curves, such as wild ramification. Furthermore, explicit formulas for the canonical height of twisted Fermat curves are obtained, implying explicit Parshin type bounds for the Arakelov metric., Comment: 61 pages

We train a model to generate images from multimodal prompts of interleaved text and images such as "a man and his dog in an animated style." We bootstrap a multimodal dataset by extracting semantically meaningful image crops corresponding to words in the image captions of synthetically generated and publicly available text-image data. Our model, MUMU, is composed of a vision-language model encoder with a diffusion decoder and is trained on a single 8xH100 GPU node. Despite being only trained on crops from the same image, MUMU learns to compose inputs from different images into a coherent output. For example, an input of a realistic person and a cartoon will output the same person in the cartoon style, and an input of a standing subject and a scooter will output the subject riding the scooter. As a result, our model generalizes to tasks such as style transfer and character consistency. Our results show the promise of using multimodal models as general purpose controllers for image generation.

One of the fundamental representation learning tasks is unsupervised sequential disentanglement, where latent codes of inputs are decomposed to a single static factor and a sequence of dynamic factors. To extract this latent information, existing methods condition the static and dynamic codes on the entire input sequence. Unfortunately, these models often suffer from information leakage, i.e., the dynamic vectors encode both static and dynamic information, or vice versa, leading to a non-disentangled representation. Attempts to alleviate this problem via reducing the dynamic dimension and auxiliary loss terms gain only partial success. Instead, we propose a novel and simple architecture that mitigates information leakage by offering a simple and effective subtraction inductive bias while conditioning on a single sample. Remarkably, the resulting variational framework is simpler in terms of required loss terms, hyperparameters, and data augmentation. We evaluate our method on multiple data-modality benchmarks including general time series, video, and audio, and we show beyond state-of-the-art results on generation and prediction tasks in comparison to several strong baselines., Comment: Accepted to ICML 2024; The first four authors contributed equally

Recent developments in static patch holography proposed that quantum gravity in de Sitter space admits a dual description in terms of a quantum mechanical theory living on a timelike surface near the cosmological horizon. In parallel, geometric observables associated with the Einstein-Rosen bridge of a black hole background were suggested to compute the computational complexity of the state dual to a gravitational theory. In this work, we pursue the study of the complexity=volume and complexity=action conjectures in a Schwarzschild-de Sitter geometry perturbed by the insertion of a shockwave at finite boundary times. This analysis extends previous studies that focused either on the complexity=volume 2.0 conjecture, or on the case of a shockwave inserted along the cosmological horizon. We show that the switchback effect, describing the delay in the evolution of complexity in reaction to a perturbation, is a universal feature of the complexity proposals in asymptotically de Sitter space. The geometric origin of this phenomenon is related to the causal connection between the static patches of de Sitter space when a positive pulse of null energy is inserted in the geometry., Comment: 42 pages+ appendices, 22 figures; v2: references added, minor changes

Most approaches for assessing causality in complex dynamical systems fail when the interactions between variables are inherently non-linear and non-stationary. Here we introduce Temporal Autoencoders for Causal Inference (TACI), a methodology that combines a new surrogate data metric for assessing causal interactions with a novel two-headed machine learning architecture to identify and measure the direction and strength of time-varying causal interactions. Through tests on both synthetic and real-world datasets, we demonstrate TACI's ability to accurately quantify dynamic causal interactions across a variety of systems. Our findings display the method's effectiveness compared to existing approaches and also highlight our approach's potential to build a deeper understanding of the mechanisms that underlie time-varying interactions in physical and biological systems.

We bootstrap the Veneziano superstring amplitude in 10 dimensions from the bottom-up. Starting with the most general maximally supersymmetric Yang-Mills EFT, we input information about the lowest-lying massive states, which we assume contribute via tree-level exchanges to the 4-point amplitude. We show the following: (1) if there is only a single state at the lowest mass, it must be a scalar. (2) Assuming a string-inspired gap between the mass of this scalar and any other massive states, the allowed region of Wilson coefficients has a new sharp corner where the Veneziano amplitude is located. (3) Upon fixing the next massive state to be a vector, the EFT bounds have a one-parameter family of corners; these would correspond to models with linear Regge trajectories of varying slopes, one of which is the open superstring. (4) When the ratio between the massive scalar coupling and the $\text{tr}\, F^4$ coefficient is fixed to its string value, the spin and mass of the second massive state is determined by the bootstrap and the Veneziano amplitude is isolated on a small island in parameter space. Finally, we compare with other recent bootstraps approaches, both the pion model and imposing Regge-inspired maximal spin constraints., Comment: 37 pages + appendices, 21 figures

We studied the effects arising from a coherent source of photons on the entanglement between excitons in a strained graphene monolayer. The graphene layer was considered to be embedded in an imperfect optical microcavity. In our investigation, we have studied the entanglement dynamics of systems consisting of up to five excitons, which are treated as atomic-like qubits. Entangled states of multiple qubits are useful in quantum error correction codes. We have monitored the time evolution of the concurrence, three-$\pi$, mutual information, and especially the negativity. We have demonstrated that coherent pumping can create lasting entanglement between the excitons. However, the entanglement only persists when the rate at which photons are pumped is smaller than the decay rate of the cavity. Our results show that the degree in entanglement between the excitons is increased with the intensity of the strain-induced pseudomagnetic field in the graphene sheet. Additionally, we have shown that a maximum amount of entanglement occurs at a finite number of excitons in the system which depends on the parameters describing the structure.

A common assumption in radar remote sensing studies for vegetation is that radar returns originate from a target made up by a set of uniformly distributed isotropic scatterers. Nonetheless, several studies in the literature have noted that orientation effects and heterogeneities have a noticeable impact in backscattering signatures according to the specific vegetation type and sensor frequency. In this paper we have employed the subaperture decomposition technique (i.e. a time-frequency analysis) and the 3-D Barakat degree of polarisation to assess the variation of the volume backscatterig power as a function of the azimuth look angle. Three different datasets, i.e. multi-frequency indoor acquisitions over short vegetation samples, and P-band airborne data and L-band satellite data over boreal and tropical forest, respectively, have been employed in this study. We have argued that despite depolarising effects may be only sensed through a small portion of the synthetic aperture, they can lead to overestimated retrievals of the volume scattering for the full resolution image. This has direct implications in the existing model-based and model-free polarimetric SAR decompositions.

The study of non-line-of-sight (NLOS) imaging is growing due to its many potential applications, including rescue operations and pedestrian detection by self-driving cars. However, implementing NLOS imaging on a moving camera remains an open area of research. Existing NLOS imaging methods rely on time-resolved detectors and laser configurations that require precise optical alignment, making it difficult to deploy them in dynamic environments. This work proposes a data-driven approach to NLOS imaging, PathFinder, that can be used with a standard RGB camera mounted on a small, power-constrained mobile robot, such as an aerial drone. Our experimental pipeline is designed to accurately estimate the 2D trajectory of a person who moves in a Manhattan-world environment while remaining hidden from the camera's field-of-view. We introduce a novel approach to process a sequence of dynamic successive frames in a line-of-sight (LOS) video using an attention-based neural network that performs inference in real-time. The method also includes a preprocessing selection metric that analyzes images from a moving camera which contain multiple vertical planar surfaces, such as walls and building facades, and extracts planes that return maximum NLOS information. We validate the approach on in-the-wild scenes using a drone for video capture, thus demonstrating low-cost NLOS imaging in dynamic capture environments., Comment: First two authors have equal contribution

We present an operations-ready multi-model ensemble weather forecasting system which uses hybrid data-driven weather prediction models coupled with the European Centre for Medium-range Weather Forecasts (ECMWF) ocean model to predict global weather at 1-degree resolution for 4 weeks of lead time. For predictions of 2-meter temperature, our ensemble on average outperforms the raw ECMWF extended-range ensemble by 4-17%, depending on the lead time. However, after applying statistical bias corrections, the ECMWF ensemble is about 3% better at 4 weeks. For other surface parameters, our ensemble is also within a few percentage points of ECMWF's ensemble. We demonstrate that it is possible to achieve near-state-of-the-art subseasonal-to-seasonal forecasts using a multi-model ensembling approach with data-driven weather prediction models.