Two competing paradigms exist for self-supervised learning of data representations.

Joint-Embedding Predictive Architecture (JEPA) is increasingly used for visual representation learning and as a component in model-based RL, but its behavior remains poorly understood. We provide a theoretical characterization of a simple, practical JEPA variant that has an auxiliary regression head trained jointly with latent dynamics. We prove a No Unhealthy Representation Collapse theorem: in deterministic MDPs, if training drives both the latent-transition consistency loss and the auxiliary regression loss to zero, then any pair of non-equivalent observations, i.e., those that do not have the same transition dynamics or auxiliary value, must map to distinct latent representations. Thus, the auxiliary task anchors which distinctions the representation must preserve. Controlled ablations in a counting environment corroborate the theory and show that training the JEPA model jointly with the auxiliary head generates a richer representation than training them separately. Our work indicates a path to improve JEPA encoders: training them with an auxiliary function that, together with the transition dynamics, encodes the right equivalence relations.

Two competing paradigms exist for self-supervised learning of data representations.

Self-supervised learning has seen great success recently in unsupervised representation learning, enabling breakthroughs in natural language and image processing. However, these methods often rely on autoregressive and masked modeling, which aim to reproduce masked information in the input, which can be vulnerable to the presence of noise or confounding variables. To address this problem, Joint-Embedding Predictive Architectures (JEPA) has been introduced with the aim to perform self-supervised learning in the latent space. To leverage these advancements in the domain of time series, we introduce Time Series JEPA (TS-JEPA), an architecture specifically adapted for time series representation learning. We validate TS-JEPA on both classification and forecasting, showing that it can match or surpass current state-of-the-art baselines on different standard datasets. Notably, our approach demonstrates a strong performance balance across diverse tasks, indicating its potential as a robust foundation for learning general representations. Thus, this work lays the groundwork for developing future time series foundation models based on Joint Embedding.

In this work, we aim to optimize the radio resource management of a communication system between a remote controller and its device, whose state is represented through image frames, without compromising the performance of the control task. We propose a novel machine learning (ML) technique to jointly model and predict the dynamics of the control system as well as the wireless propagation environment in latent space. Our method leverages two coupled joint-embedding predictive architectures (JEP As): a control JEP A models the control dynamics and guides the predictions of a wireless JEP A, which captures the dynamics of the device's channel state information (CSI) through cross-modal conditioning. We then train a deep reinforcement learning (RL) algorithm to derive a control policy from latent control dynamics and a power predictor to estimate scheduling intervals with favorable channel conditions based on latent CSI representations. As such, the controller minimizes the usage of radio resources by utilizing the coupled JEP A networks to imagine the device's trajectory in latent space. We present simulation results on synthetic multimodal data and show that our proposed approach reduces transmit power by over 50% while maintaining control performance comparable to baseline methods that do not account for wireless optimization.

Traditional time series models are task-specific and often depend on dataset-specific training and extensive feature engineering. While Transformer-based architectures have improved scalability, foundation models, commonplace in text, vision, and audio, remain under-explored for time series and are largely restricted to forecasting. We introduce $\textbf{CHARM}$, a foundation embedding model for multivariate time series that learns shared, transferable, and domain-aware representations. To address the unique difficulties of time series foundation learning, $\textbf{CHARM}$ incorporates architectural innovations that integrate channel-level textual descriptions while remaining invariant to channel order. The model is trained using a Joint Embedding Predictive Architecture (JEPA), with novel augmentation schemes and a loss function designed to improve interpretability and training stability. Our $7$M-parameter model achieves state-of-the-art performance across diverse downstream tasks, setting a new benchmark for time series representation learning.

Case 03 of the Special Jurisdiction for Peace (JEP), focused on the so-called false positives in Colombia, represents one of the most harrowing episodes of the Colombian armed conflict. This article proposes an innovative methodology based on natural language analysis and semantic co-occurrence models to explore, systematize, and visualize narrative patterns present in the public hearings of victims and appearing parties. By constructing skipgram networks and analyzing their modularity, the study identifies thematic clusters that reveal regional and procedural status differences, providing empirical evidence on dynamics of victimization, responsibility, and acknowledgment in this case. This computational approach contributes to the collective construction of both judicial and extrajudicial truth, offering replicable tools for other transitional justice cases. The work is grounded in the pillars of truth, justice, reparation, and non-repetition, proposing a critical and in-depth reading of contested memories.

Visual Analytics (VA) tools and techniques have been instrumental in supporting users to build better classification models, interpret models' overall logic, and audit results. In a different direction, VA has recently been applied to transform classification models into descriptive mechanisms instead of predictive. The idea is to use such models as surrogates for data patterns, visualizing the model to understand the phenomenon represented by the data. Although very useful and inspiring, the few proposed approaches have opted to use low complex classification models to promote straightforward interpretation, presenting limitations to capture intricate data patterns. In this paper, we present VAX (multiVariate dAta eXplanation), a new VA method to support the identification and visual interpretation of patterns in multivariate datasets. Unlike the existing similar approaches, VAX uses the concept of Jumping Emerging Patterns to identify and aggregate several diversified patterns, producing explanations through logic combinations of data variables. The potential of VAX to interpret complex multivariate datasets is demonstrated through use cases employing two real-world datasets covering different scenarios.

Bogota - Carmenza Gomez was planning a surprise Christmas dinner in the winter of 2008 to celebrate having her eight children back together under one roof in their home in an impoverished suburb in Bogota, the capital of Colombia. That summer, the family had finally been reunited after years apart due to the sons' military service. It was months away, but Carmenza wanted to throw an elaborate dinner to share their first Christmas together in years. But just days after the last of her sons arrived home, 23-year-old Victor Fernando, her third youngest, disappeared. "I didn't tell any of them what I was planning [for Christmas]," Carmenza recalled nearly a decade later.