arXiv subjects

Javier Flores

Publications and source records attributed to Javier Flores.

7 recordsLinked to original sources

arXiv2026

Towards High-Throughput Visible Photonic Lanterns for the EMARCOT Project

Photonic EMARCOT is an innovative project involving Spanish, German and Australian research institutes that aims to integrate multiple Optical Tube Assemblies (OTAs) using photonic lanterns. The "Pathfinder" prototype, featuring seven OTAs with a 1.1-meter effective aperture, will feed a spectrograph at the Calar Alto Observatory, with first light expected in 2026. We report the fabrication and experimental evaluation of a custom 7x1 multi-mode photonic lantern (MMPL) developed for this framework, featuring seven 25 um core multi-mode inputs merging into a single 50 um core multi-mode output optimized for the visible wavelength range (400-700 nm). Optical characterization centered at 600 nm reveals exceptional channel-to-channel uniformity, with statistical variations close to zero across both bare-fiber and connectorized MMPL configurations. However, the total baseline throughput of this initial device was limited to below 4%. From the refractive index studies, this low throughput is attributed to severe refractive index mismatch between the internal fiber cladding geometry and the structural capillary, which suppresses total internal reflection during the tapering transition. This work establishes an important diagnostic baseline that highlights the necessary fabrication tolerances needed to improve future high-throughput manufacturing processes for precision radial-velocity astronomy.

astro-ph.IM↗

arXiv2026

Towards a Comprehensive Understanding of Planetary Systems through Population-Level, Large-Scale Surveys

Over the past three decades, exoplanet research has delivered an extensive census of planets spanning a wide range of masses, sizes, and orbital configurations. Despite this progress, the physical interpretation of these populations remains severely limited, as precise constraints on planetary masses, interior structures, and atmospheres are available only for a small, highly selected subset of targets. As a result, most known exoplanets remain physically ambiguous, preventing the construction of robust population-level trends and limiting our understanding of planet formation, evolution, and habitability. In the coming decades, missions such as PLATO, Earth 2.0, and the Nancy Grace Roman Space Telescope will dramatically expand the number of exoplanets detected. However, without a corresponding capability to characterise planetary masses and atmospheres at scale, these discoveries will remain largely detection-driven. Current and planned facilities, including JWST and ELT-class instruments, excel at detailed studies of individual systems but are intrinsically unsuited for large, homogeneous surveys. This white paper identifies population-level physical characterisation as a fundamental science challenge for the 2040s and motivates the need for a new observational paradigm. We outline how photonics-enabled, modular telescope architectures can deliver the survey speed, stability, and scalability required to jointly probe planetary interiors and atmospheres across statistically meaningful samples, thereby enabling a comprehensive and physically grounded understanding of planetary systems.

astro-ph.IM↗

arXiv2025

Which Way from B to A: The role of embedding geometry in image interpolation for Stable Diffusion

It can be shown that Stable Diffusion has a permutation-invariance property with respect to the rows of Contrastive Language-Image Pretraining (CLIP) embedding matrices. This inspired the novel observation that these embeddings can naturally be interpreted as point clouds in a Wasserstein space rather than as matrices in a Euclidean space. This perspective opens up new possibilities for understanding the geometry of embedding space. For example, when interpolating between embeddings of two distinct prompts, we propose reframing the interpolation problem as an optimal transport problem. By solving this optimal transport problem, we compute a shortest path (or geodesic) between embeddings that captures a more natural and geometrically smooth transition through the embedding space. This results in smoother and more coherent intermediate (interpolated) images when rendered by the Stable Diffusion generative model. We conduct experiments to investigate this effect, comparing the quality of interpolated images produced using optimal transport to those generated by other standard interpolation methods. The novel optimal transport--based approach presented indeed gives smoother image interpolations, suggesting that viewing the embeddings as point clouds (rather than as matrices) better reflects and leverages the geometry of the embedding space.

cs.CV↗

arXiv2025

Consistency of Feature Attribution in Deep Learning Architectures for Multi-Omics

Machine and deep learning have grown in popularity and use in biological research over the last decade but still present challenges in interpretability of the fitted model. The development and use of metrics to determine features driving predictions and increase model interpretability continues to be an open area of research. We investigate the use of Shapley Additive Explanations (SHAP) on a multi-view deep learning model applied to multi-omics data for the purposes of identifying biomolecules of interest. Rankings of features via these attribution methods are compared across various architectures to evaluate consistency of the method. We perform multiple computational experiments to assess the robustness of SHAP and investigate modeling approaches and diagnostics to increase and measure the reliability of the identification of important features. Accuracy of a random-forest model fit on subsets of features selected as being most influential as well as clustering quality using only these features are used as a measure of effectiveness of the attribution method. Our findings indicate that the rankings of features resulting from SHAP are sensitive to the choice of architecture as well as different random initializations of weights, suggesting caution when using attribution methods on multi-view deep learning models applied to multi-omics data. We present an alternative, simple method to assess the robustness of identification of important biomolecules.

stat.ML↗

arXiv2024

FREYJA: Efficient Join Discovery in Data Lakes

Data lakes are massive repositories of raw and heterogeneous data, designed to meet the requirements of modern data storage. Nonetheless, this same philosophy increases the complexity of performing discovery tasks to find relevant data for subsequent processing. As a response to these growing challenges, we present FREYJA, a modern data discovery system capable of effectively exploring data lakes, aimed at finding candidates to perform joins and increase the number of attributes for downstream tasks. More precisely, we want to compute rankings that sort potential joins by their relevance. Modern mechanisms apply advanced table representation learning (TRL) techniques to yield accurate joins. Yet, this incurs high computational costs when dealing with elevated volumes of data. In contrast to the state-of-the-art, we adopt a novel notion of join quality tailored to data lakes, which leverages syntactic measurements while achieving accuracy comparable to that of TRL approaches. To obtain this metric in a scalable manner we train a general purpose predictive model. Predictions are based, rather than on large-scale datasets, on data profiles, succinct representations that capture the underlying characteristics of the data. Our experiments show that our system, FREYJA, matches the results of the state-of-the-art whilst reducing the execution times by several orders of magnitude.

cs.DB↗

arXiv2023

Measuring and Predicting the Quality of a Join for Data Discovery

We study the problem of discovering joinable datasets at scale. We approach the problem from a learning perspective relying on profiles. These are succinct representations that capture the underlying characteristics of the schemata and data values of datasets, which can be efficiently extracted in a distributed and parallel fashion. Profiles are then compared, to predict the quality of a join operation among a pair of attributes from different datasets. In contrast to the state-of-the-art, we define a novel notion of join quality that relies on a metric considering both the containment and cardinality proportion between join candidate attributes. We implement our approach in a system called NextiaJD, and present experiments to show the predictive performance and computational efficiency of our method. Our experiments show that NextiaJD obtains greater predictive performance to that of hash-based methods while we are able to scale-up to larger volumes of data.

cs.DB↗

arXiv2020

Scalable Data Discovery Using Profiles

We study the problem of discovering joinable datasets at scale. This is, how to automatically discover pairs of attributes in a massive collection of independent, heterogeneous datasets that can be joined. Exact (e.g., based on distinct values) and hash-based (e.g., based on locality-sensitive hashing) techniques require indexing the entire dataset, which is unattainable at scale. To overcome this issue, we approach the problem from a learning perspective relying on profiles. These are succinct representations that capture the underlying characteristics of the schemata and data values of datasets, which can be efficiently extracted in a distributed and parallel fashion. Profiles are then compared, to predict the quality of a join operation among a pair of attributes from different datasets. In contrast to the state-of-the-art, we define a novel notion of join quality that relies on a metric considering both the containment and cardinality proportions between candidate attributes. We implement our approach in a system called NextiaJD, and present extensive experiments to show the predictive performance and computational efficiency of our method. Our experiments show that NextiaJD obtains similar predictive performance to that of hash-based methods, yet we are able to scale-up to larger volumes of data. Also, NextiaJD generates a considerably less amount of false positives, which is a desirable feature at scale.

cs.DB↗