Search

Latent Video Diffusion Models can easily deceive casual observers and domain experts alike thanks to the produced image quality and temporal consistency. Beyond entertainment, this creates opportunities around safe data sharing of fully synthetic datasets, which are crucial in healthcare, as well as other domains relying on sensitive personal information. However, privacy concerns with this approach have not fully been addressed yet, and models trained on synthetic data for specific downstream tasks still perform worse than those trained on real data. This discrepancy may be partly due to the sampling space being a subspace of the training videos, effectively reducing the training data size for downstream models. Additionally, the reduced temporal consistency when generating long videos could be a contributing factor. In this paper, we first show that training privacy-preserving models in latent space is computationally more efficient and generalize better. Furthermore, to investigate downstream degradation factors, we propose to use a re-identification model, previously employed as a privacy preservation filter. We demonstrate that it is sufficient to train this model on the latent space of the video generator. Subsequently, we use these models to evaluate the subspace covered by synthetic video datasets and thus introduce a new way to measure the faithfulness of generative machine learning models. We focus on a specific application in healthcare echocardiography to illustrate the effectiveness of our novel methods. Our findings indicate that only up to 30.8% of the training videos are learned in latent video diffusion models, which could explain the lack of performance when training downstream tasks on synthetic data., Comment: 8 pages, 5 tables, 6 figures

The Dzyaloshinskii-Lifshitz-Pitaevskii (DLP) theory of Casimir forces predicts a repulsion between two material surfaces separated by a third medium with an intermediate dielectric function. This DLP repulsion paradigm constitutes an important example with many applications. We show here that it is broken when the surfaces interact across salted water due to the nonlocal response of the ions in solution. We consider the interaction between silica and gold and show that the zero-frequency contribution, which is the only one affected by the nonlocal response, is dominant for distances in the range near $100\,{\rm nm}$ and beyond. As a result, the total Casimir force between gold and silica surfaces in salted water is always attractive in spite of the DLP repulsion paradigm., Comment: 9 pages, 4 figures

Human-robot interaction through mixed reality (MR) technologies enables novel, intuitive interfaces to control robots in remote operations. Such interfaces facilitate operations in hazardous environments, where human presence is risky, yet human oversight remains crucial. Potential environments include disaster response scenarios and areas with high radiation or toxic chemicals. In this paper we present an interface system projecting a 3D representation of a scanned room as a scaled-down 'dollhouse' hologram, allowing users to select and manipulate objects using a straightforward drag-and-drop interface. We then translate these drag-and-drop user commands into real-time robot actions based on the recent Spot-Compose framework. The Unity-based application provides an interactive tutorial and a user-friendly experience, ensuring ease of use. Through comprehensive end-to-end testing, we validate the system's capability in executing pick-and-place tasks and a complementary user study affirms the interface's intuitive controls. Our findings highlight the advantages of this interface in improving user experience and operational efficiency. This work lays the groundwork for a robust framework that advances the potential for seamless human-robot collaboration in diverse applications. Paper website: https://holospot.github.io/, Comment: 6 pages, 8 figures, submitted to ICRA 2025

Although powerful for image generation, consistent and controllable video is a longstanding problem for diffusion models. Video models require extensive training and computational resources, leading to high costs and large environmental impacts. Moreover, video models currently offer limited control of the output motion. This paper introduces a novel approach to video generation by augmenting image diffusion models to create sequential animation frames while maintaining fine detail. These techniques can be applied to existing image models without training any video parameters (zero-shot) by altering the input noise in a latent diffusion model. Two complementary methods are presented. Noise crystallization ensures consistency but is limited to large movements due to reduced latent embedding sizes. Liquid noise trades consistency for greater flexibility without resolution limitations. The core concepts also allow other applications such as relighting, seamless upscaling, and improved video style transfer. Furthermore, an exploration of the VAE embedding used for latent diffusion models is performed, resulting in interesting theoretical insights such as a method for human-interpretable latent spaces.

We introduce the Joint Video-Image Diffusion model (JVID), a novel approach to generating high-quality and temporally coherent videos. We achieve this by integrating two diffusion models: a Latent Image Diffusion Model (LIDM) trained on images and a Latent Video Diffusion Model (LVDM) trained on video data. Our method combines these models in the reverse diffusion process, where the LIDM enhances image quality and the LVDM ensures temporal consistency. This unique combination allows us to effectively handle the complex spatio-temporal dynamics in video generation. Our results demonstrate quantitative and qualitative improvements in producing realistic and coherent videos.

A low energy particle confined by a horizontal reflective surface and gravity settles in gravitationally bound quantum states. These gravitational quantum states (GQS) were so far only observed with neutrons. However, the existence of GQS is predicted also for atoms. The GRASIAN collaboration pursues the first observation of GQS of atoms, using a cryogenic hydrogen beam. This endeavor is motivated by the higher densities, which can be expected from hydrogen compared to neutrons, the easier access, the fact that GQS were never observed with atoms and the accessibility to hypothetical short range interactions. In addition to enabling gravitational quantum spectroscopy, such a cryogenic hydrogen beam with very low vertical velocity components - a few cm s$^{-1}$, can be used for precision optical and microwave spectroscopy. In this article, we report on our methods developed to reduce background and to detect atoms with a low horizontal velocity, which are needed for such an experiment. Our recent measurement results on the collimation of the hydrogen beam to 2 mm, the reduction of background and improvement of signal-to-noise and finally our first detection of atoms with velocities < 72 m s$^{-1}$ are presented. Furthermore, we show calculations, estimating the feasibility of the planned experiment and simulations which confirm that we can select vertical velocity components in the order of cm s$^{-1}$.

Diagnosing medical conditions from histopathology data requires a thorough analysis across the various resolutions of Whole Slide Images (WSI). However, existing generative methods fail to consistently represent the hierarchical structure of WSIs due to a focus on high-fidelity patches. To tackle this, we propose Ultra-Resolution Cascaded Diffusion Models (URCDMs) which are capable of synthesising entire histopathology images at high resolutions whilst authentically capturing the details of both the underlying anatomy and pathology at all magnification levels. We evaluate our method on three separate datasets, consisting of brain, breast and kidney tissue, and surpass existing state-of-the-art multi-resolution models. Furthermore, an expert evaluation study was conducted, demonstrating that URCDMs consistently generate outputs across various resolutions that trained evaluators cannot distinguish from real images. All code and additional examples can be found on GitHub., Comment: arXiv admin note: text overlap with arXiv:2312.01152

To make medical datasets accessible without sharing sensitive patient information, we introduce a novel end-to-end approach for generative de-identification of dynamic medical imaging data. Until now, generative methods have faced constraints in terms of fidelity, spatio-temporal coherence, and the length of generation, failing to capture the complete details of dataset distributions. We present a model designed to produce high-fidelity, long and complete data samples with near-real-time efficiency and explore our approach on a challenging task: generating echocardiogram videos. We develop our generation method based on diffusion models and introduce a protocol for medical video dataset anonymization. As an exemplar, we present EchoNet-Synthetic, a fully synthetic, privacy-compliant echocardiogram dataset with paired ejection fraction labels. As part of our de-identification protocol, we evaluate the quality of the generated dataset and propose to use clinical downstream tasks as a measurement on top of widely used but potentially biased image quality metrics. Experimental outcomes demonstrate that EchoNet-Synthetic achieves comparable dataset fidelity to the actual dataset, effectively supporting the ejection fraction regression task. Code, weights and dataset are available at https://github.com/HReynaud/EchoNet-Synthetic., Comment: Accepted at MICCAI 2024

The present work aims at proving mathematically that a neural network inspired by biology can learn a classification task thanks to local transformations only. In this purpose, we propose a spiking neural network named CHANI (Correlation-based Hawkes Aggregation of Neurons with bio-Inspiration), whose neurons activity is modeled by Hawkes processes. Synaptic weights are updated thanks to an expert aggregation algorithm, providing a local and simple learning rule. We were able to prove that our network can learn on average and asymptotically. Moreover, we demonstrated that it automatically produces neuronal assemblies in the sense that the network can encode several classes and that a same neuron in the intermediate layers might be activated by more than one class, and we provided numerical simulations on synthetic dataset. This theoretical approach contrasts with the traditional empirical validation of biologically inspired networks and paves the way for understanding how local learning rules enable neurons to form assemblies able to represent complex concepts.

We prove oracle inequalities for a penalized log-likelihood criterion that hold even if the data are not independent and not stationary, based on a martingale approach. The assumptions are checked for various contexts: density estimation with independent and identically distributed (i.i.d) data, hidden Markov models, spiking neural networks, adversarial bandits. In each case, we compare our results to the literature, showing that, although we lose some logarithmic factors in the most classical case (i.i.d.), these results are comparable or more general than the existing results in the more dependent cases.

By effectively implementing the strategies for resource allocation, the capabilities, and reliability of non-terrestrial networks (NTN) can be enhanced. This leads to enhance spectrum utilization performance while minimizing the unmet system capacity, meeting quality of service (QoS) requirements and overall system optimization. In turn, a wide range of applications and services in various domains can be supported. However, allocating resources in a multi-constellation system with heterogeneous satellite links and highly dynamic user traffic demand pose challenges in ensuring sufficient and fair resource distribution. To mitigate these complexities and minimize the overhead, there is a growing shift towards utilizing artificial intelligence (AI) for its ability to handle such problems effectively. This calls for the development of an intelligent decision-making controller using AI to efficiently manage resources in this complex environment. In this context, real-world open datasets play a pivotal role in the development of AI models addressing radio control optimization problems. As a matter of fact, acquiring suitable datasets can be arduous. Therefore, this paper identifies pertinent real-world open datasets representing realistic traffic pattern, network performances and demand for fixed and dynamic user terminals, enabling a variety of uses cases. The aim of gathering and publishing the information of these datasets are to inspire and assist the research community in crafting the advance resource management solutions. In a nutshell, this paper establishes a solid foundation of commercially accessible data, with the potential to set benchmarks and accelerate the resolution of resource allocation optimization challenges., Comment: In the proceedings of IEEE Future Networks World Forum 13_15 November 2023, Baltimore, MD, USA

We develop and study a statistical test to detect synchrony in spike trains. Our test is based on the number of coincidences between two trains of spikes. The data are supplied in the form of \(n\) pairs (assumed to be independent) of spike trains. The aim is to assess whether the two trains in a pair are also independent. Our approach is based on previous results of Albert et al. (2015, 2019) and Kim et al. (2022) that we extend to our setting, focusing on the construction of a non-asymptotic criterion ensuring the detection of synchronization in the framework of permutation tests. Our criterion is constructed such that it ensures the control of the Type II error, while the Type I error is controlled by construction. We illustrate our results within two classical models of interacting neurons, the jittering Poisson model and Hawkes processes having \(M\) components interacting in a mean field frame and evolving in stationary regime. For this latter model, we obtain a lower bound of the size \(n\) of the sample necessary to detect the dependency between two neurons.

To address the major issues associated with using Time-Use Survey (TUS) for simulating residential load curves, we present the SMACH approach, which combines qualitative and quantitative data with agent-based simulation. Our model consists of autonomous agents assigned with daily tasks. The agents try to accomplish their assigned tasks to the best of their abilities. Quantitative data are used to generate tasks assignments. Qualitative studies allow us to define how agents select, based on plausible cognitive principles, the tasks to accomplish depending on the context. Our results show a better representation of weekdays and weekends, a more flexible association of tasks with appliances, and an improved simulation of load curves compared to real data. Highlights $\bullet$ Discussion about Time-Use Surveys (TUS) limits and the use of TUS in activity and energy simulation $\bullet$ Presentation of complementary data both qualitative and quantitative used to complement TUS data $\bullet$ Proposition of an agent-based approach that balances these limitations

Diagnoses from histopathology images rely on information from both high and low resolutions of Whole Slide Images. Ultra-Resolution Cascaded Diffusion Models (URCDMs) allow for the synthesis of high-resolution images that are realistic at all magnification levels, focusing not only on fidelity but also on long-distance spatial coherency. Our model beats existing methods, improving the pFID-50k [2] score by 110.63 to 39.52 pFID-50k. Additionally, a human expert evaluation study was performed, reaching a weighted Mean Absolute Error (MAE) of 0.11 for the Lower Resolution Diffusion Models and a weighted MAE of 0.22 for the URCDM., Comment: MedNeurIPS 2023 poster

This work presents an extensive hyperparameter search on Image Diffusion Models for Echocardiogram generation. The objective is to establish foundational benchmarks and provide guidelines within the realm of ultrasound image and video generation. This study builds over the latest advancements, including cutting-edge model architectures and training methodologies. We also examine the distribution shift between real and generated samples and consider potential solutions, crucial to train efficient models on generated data. We determine an Optimal FID score of $0.88$ for our research problem and achieve an FID of $2.60$. This work is aimed at contributing valuable insights and serving as a reference for further developments in the specialized field of ultrasound image and video generation., Comment: MedNeurIPS 2023 poster

Photonics offer numerous functionalities that can be used to realize astrophotonic instruments. The most spectacular example to date is the ESO Gravity instrument at the Very Large Telescope in Chile. Integrated astrophotonic devices stand to offer critical advantages for instrument development, including extreme miniaturization, as well as integration, superior thermal and mechanical stabilization owing to the small footprint, and high replicability offering cost savings. Numerous astrophotonic technologies have been developed to address shortcomings of conventional instruments to date, including for example the development of photonic lanterns, complex aperiodic fiber Bragg gratings, complex beam combiners to enable long baseline interferometry, and laser frequency combs for high precision spectral calibration of spectrometers. Despite these successes, the facility implementation of photonic solutions in astronomical instrumentation is currently limited because of (1) low throughputs from coupling to fibers, coupling fibers to chips, propagation and bend losses, device losses, etc, (2) difficulties with scaling to large channel count devices needed for large bandwidths and high resolutions, and (3) efficient integration of photonics with detectors, to name a few. In this roadmap, we identify 24 areas that need further development. We outline the challenges and advances needed across those areas covering design tools, simulation capabilities, fabrication processes, the need for entirely new components, integration and hybridization and the characterization of devices. To realize these advances the astrophotonics community will have to work cooperatively with industrial partners who have more advanced manufacturing capabilities. With the advances described herein, multi-functional instruments will be realized leading to novel observing capabilities for both ground and space platforms., Comment: 191 pages, 47 figures. This is the version of the article before peer review or editing, as submitted by an author to J. Phys. Photonics. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at https://iopscience.iop.org/article/10.1088/2515-7647/ace869/meta

This document presents a summary of the 2023 Terrestrial Very-Long-Baseline Atom Interferometry Workshop hosted by CERN. The workshop brought together experts from around the world to discuss the exciting developments in large-scale atom interferometer (AI) prototypes and their potential for detecting ultralight dark matter and gravitational waves. The primary objective of the workshop was to lay the groundwork for an international TVLBAI proto-collaboration. This collaboration aims to unite researchers from different institutions to strategize and secure funding for terrestrial large-scale AI projects. The ultimate goal is to create a roadmap detailing the design and technology choices for one or more km-scale detectors, which will be operational in the mid-2030s. The key sections of this report present the physics case and technical challenges, together with a comprehensive overview of the discussions at the workshop together with the main conclusions., Comment: Summary of the Terrestrial Very-Long-Baseline Atom Interferometry Workshop held at CERN: https://indico.cern.ch/event/1208783/

We develop several statistical tests of the determinant of the diffusion coefficient of a stochastic differential equation, based on discrete observations on a time interval $[0,T]$ sampled with a time step $\Delta$. Our main contribution is to control the test Type I and Type II errors in a non asymptotic setting, i.e. when the number of observations and the time step are fixed. The test statistics are calculated from the process increments. In dimension 1, the density of the test statistic is explicit. In dimension 2, the test statistic has no explicit density but upper and lower bounds are proved. We also propose a multiple testing procedure in dimension greater than 2. Every test is proved to be of a given non-asymptotic level and separability conditions to control their power are also provided. A numerical study illustrates the properties of the tests for stochastic processes with known or estimated drifts.

We report on the first production of an antihydrogen beam by charge exchange of 6.1 keV antiprotons with a cloud of positronium in the GBAR experiment at CERN. The antiproton beam was delivered by the AD/ELENA facility. The positronium target was produced from a positron beam itself obtained from an electron linear accelerator. We observe an excess over background indicating antihydrogen production with a significance of 3-4 standard deviations.

The electromagnetic Casimir interaction between dielectric objects immersed in salted water includes a universal contribution that is not screened by the solvent and therefore long-ranged. Here, we study the geometry of two parallel dielectric cylinders. We derive the Casimir free energy by using the scattering method. We show that its magnitude largely exceeds the thermal energy scale for a large parameter range. This includes length scales relevant for actin filaments and microtubules in cells. We show that the Casimir free energy is a universal function of the geometry, independent of the dielectric response functions of the cylinders, at all distances of biological interest. While multiple interactions exist between filaments in cells, this universal attractive interaction should have an important role in the cohesion of bundles of parallel filaments., Comment: 15 pages, 9 figures. This paper extends the unpublished short report arXiv:2304.06006

GenerateCT, the first approach to generating 3D medical imaging conditioned on free-form medical text prompts, incorporates a text encoder and three key components: a novel causal vision transformer for encoding 3D CT volumes, a text-image transformer for aligning CT and text tokens, and a text-conditional super-resolution diffusion model. Without directly comparable methods in 3D medical imaging, we benchmarked GenerateCT against cutting-edge methods, demonstrating its superiority across all key metrics. Importantly, we evaluated GenerateCT's clinical applications in a multi-abnormality classification task. First, we established a baseline by training a multi-abnormality classifier on our real dataset. To further assess the model's generalization to external data and performance with unseen prompts in a zero-shot scenario, we employed an external set to train the classifier, setting an additional benchmark. We conducted two experiments in which we doubled the training datasets by synthesizing an equal number of volumes for each set using GenerateCT. The first experiment demonstrated an 11% improvement in the AP score when training the classifier jointly on real and generated volumes. The second experiment showed a 7% improvement when training on both real and generated volumes based on unseen prompts. Moreover, GenerateCT enables the scaling of synthetic training datasets to arbitrary sizes. As an example, we generated 100,000 3D CTs, fivefold the number in our real set, and trained the classifier exclusively on these synthetic CTs. Impressively, this classifier surpassed the performance of the one trained on all available real data by a margin of 8%. Last, domain experts evaluated the generated volumes, confirming a high degree of alignment with the text prompt. Access our code, model weights, training data, and generated data at https://github.com/ibrahimethemhamamci/GenerateCT

When fitting the learning data of an individual to algorithm-like learning models, the observations are so dependent and non-stationary that one may wonder what the classical Maximum Likelihood Estimator (MLE) could do, even if it is the usual tool applied to experimental cognition. Our objective in this work is to show that the estimation of the learning rate cannot be efficient if the learning rate is constant in the classical Exp3 (Exponential weights for Exploration and Exploitation) algorithm. Secondly, we show that if the learning rate decreases polynomially with the sample size, then the prediction error and in some cases the estimation error of the MLE satisfy bounds in probability that decrease at a polynomial rate.