Current experimental scientists have been increasingly relying on simulation-based inference (SBI) to invert complex non-linear models with intractable likelihoods. However, posterior approximations obtained with SBI are often miscalibrated, causing credible regions to undercover true parameters. We develop $\texttt{CP4SBI}$, a model-agnostic conformal calibration framework that constructs credible sets with local Bayesian coverage. Our two proposed variants, namely local calibration via regression trees and CDF-based calibration, enable finite-sample local coverage guarantees for any scoring function, including HPD, symmetric, and quantile-based regions. Experiments on widely used SBI benchmarks demonstrate that our approach improves the quality of uncertainty quantification for neural posterior estimators using both normalizing flows and score-diffusion modeling.

Cross-validation plays a fundamental role in Machine Learning, enabling robust evaluation of model performance and preventing overestimation on training and validation data. However, one of its drawbacks is the potential to create data subsets (folds) that do not adequately represent the diversity of the original dataset, which can lead to biased performance estimates. The objective of this work is to deepen the investigation of cluster-based cross-validation strategies by analyzing the performance of different clustering algorithms through experimental comparison. Additionally, a new cross-validation technique that combines Mini Batch K-Means with class stratification is proposed. Experiments were conducted on 20 datasets (both balanced and imbalanced) using four supervised learning algorithms, comparing cross-validation strategies in terms of bias, variance, and computational cost. The technique that uses Mini Batch K-Means with class stratification outperformed others in terms of bias and variance on balanced datasets, though it did not significantly reduce computational cost. On imbalanced datasets, traditional stratified cross-validation consistently performed better, showing lower bias, variance, and computational cost, making it a safe choice for performance evaluation in scenarios with class imbalance. In the comparison of different clustering algorithms, no single algorithm consistently stood out as superior. Overall, this work contributes to improving predictive model evaluation strategies by providing a deeper understanding of the potential of cluster-based data splitting techniques and reaffirming the effectiveness of well-established strategies like stratified cross-validation. Moreover, it highlights perspectives for increasing the robustness and reliability of model evaluations, especially in datasets with clustering characteristics.

--With the increasing availability of meteorological data from various sensors, numerical models and reanalysis products, the need for efficient data integration methods has become paramount for improving weather forecasts and hy-drometeorological studies. In this work, we propose a data fusion approach for precipitation nowcasting by integrating data from meteorological and rain gauge stations in Rio de Janeiro metropolitan area with ERA5 reanalysis data and GFS numerical weather prediction. We employ the spatiotemporal deep learning architecture called STConvS2S, leveraging a structured dataset covering a 9 x 11 grid. The study spans from January 2011 to October 2024, and we evaluate the impact of integrating three surface station systems. Among the tested configurations, the fusion-based model achieves an F1-score of 0.2033 for forecasting heavy precipitation events (greater than 25 mm/h) at a one-hour lead time. Additionally, we present an ablation study to assess the contribution of each station network and propose a refined inference strategy for precipitation nowcasting, integrating the GFS numerical weather prediction (NWP) data with in-situ observations. Precipitation nowcasting (or very short-range forecasting [1]) involves predicting rainfall within a six-hour lead time. Objective analysis techniques are then employed to synthesize these disparate measurements into a coherent, gridded spatial map for precipitation nowcasting [16]. Accurate precipitation forecasting is critical for mitigating natural disasters, such as floods, landslides, and droughts, and supports informed decision-making across sectors including agriculture, transportation, energy, and public health [3]. Recent advancements in machine learning, particularly deep learning, have demonstrated significant potential in geoscien-tific applications, including precipitation nowcasting.

Low-resource machine translation remains a significant challenge for large language models (LLMs), which often lack exposure to these languages during pretraining and have limited parallel data for fine-tuning. We propose a novel approach that enhances translation for low-resource languages by integrating an external dictionary tool and training models end-to-end using reinforcement learning, in addition to supervised fine-tuning. Focusing on the Spanish-Wayuunaiki language pair, we frame translation as a tool-augmented decision-making problem in which the model can selectively consult a bilingual dictionary during generation. Our method combines supervised instruction tuning with Guided Reward Policy Optimization (GRPO), enabling the model to learn both when and how to use the tool effectively. BLEU similarity scores are used as rewards to guide this learning process. Preliminary results show that our tool-augmented models achieve up to +3.37 BLEU improvement over previous work, and a 18% relative gain compared to a supervised baseline without dictionary access, on the Spanish-Wayuunaiki test set from the AmericasNLP 2025 Shared Task. We also conduct ablation studies to assess the effects of model architecture and training strategy, comparing Qwen2.5-0.5B-Instruct with other models such as LLaMA and a prior NLLB-based system.

The 2020s have been witnessing a very significant advance in the development of generative artificial intelligence tools, including text generation systems based on large language models. These tools have been increasingly used to generate texts in the most diverse domains -- from technical texts to literary texts --, which might eventually lead to a lower volume of written text production by humans. This article discusses the possibility of a future in which human beings will have lost or significantly decreased their ability to write due to the outsourcing of this activity to machines. This possibility parallels the loss of the ability to write in other moments of human history, such as during the so-called Greek Dark Ages (approx. 1200 BCE - 800 BCE).

An invasion of small metallic orbs has been spotted hovering over the US in recent years, leaving the Pentagon scrambling to identify these mysterious UFOs. A new report from the crowdsourced platform Enigma, which allows people to report sightings of unidentified flying objects (UFOs), reveals more than 8,000 sightings across the US between December 2022 and June 2025. Among these, 422 reports specifically describe metallic orbs, with the majority observed between 1am and 4am near military installations in New York, California, and Arizona. Eyewitnesses, including civilians, pilots, and military personnel, reported seeing the spheres hover silently before moving at extreme speeds, leaving no trace of their departure. Some of the sightings have been captured on video or radar, though many remain unexplained.

We introduce the first best-of-both-worlds algorithm for contextual combinatorial semi-bandits that simultaneously guarantees $\widetilde{\mathcal{O}}(\sqrt{T})$ regret in the adversarial regime and $\widetilde{\mathcal{O}}(\ln T)$ regret in the corrupted stochastic regime. Our approach builds on the Follow-the-Regularized-Leader (FTRL) framework equipped with a Shannon entropy regularizer, yielding a flexible method that admits efficient implementations. Beyond regret bounds, we tackle the practical bottleneck in FTRL (or, equivalently, Online Stochastic Mirror Descent) arising from the high-dimensional projection step encountered in each round of interaction. By leveraging the Karush-Kuhn-Tucker conditions, we transform the $K$-dimensional convex projection problem into a single-variable root-finding problem, dramatically accelerating each round. Empirical evaluations demonstrate that this combined strategy not only attains the attractive regret bounds of best-of-both-worlds algorithms but also delivers substantial per-round speed-ups, making it well-suited for large-scale, real-time applications.

Machine unlearning aims to remove private or sensitive data from a pre-trained model while preserving the model's robustness. Despite recent advances, this technique has not been explored in medical image classification. This work evaluates the SalUn unlearning model by conducting experiments on the PathMNIST, OrganAMNIST, and BloodMNIST datasets. W e also analyze the impact of data augmentation on the quality of unlearning. Results show that SalUn achieves performance close to full retraining, indicating an efficient solution for use in medical applications.

This paper explores the combination of two intrinsic motivation strategies to improve the efficiency of reinforcement learning (RL) agents in environments with extreme sparse rewards, where traditional learning struggles due to infrequent positive feedback. We propose integrating Variational State as Intrinsic Reward (VSIMR), which uses Variational AutoEncoders (VAEs) to reward state novelty, with an intrinsic reward approach derived from Large Language Models (LLMs). The LLMs leverage their pre-trained knowledge to generate reward signals based on environment and goal descriptions, guiding the agent. We implemented this combined approach with an Actor-Critic (A2C) agent in the MiniGrid DoorKey environment, a benchmark for sparse rewards. Our empirical results show that this combined strategy significantly increases agent performance and sampling efficiency compared to using each strategy individually or a standard A2C agent, which failed to learn. Analysis of learning curves indicates that the combination effectively complements different aspects of the environment and task: VSIMR drives exploration of new states, while the LLM-derived rewards facilitate progressive exploitation towards goals.

We propose a novel Bayesian nonparametric classification model that combines a Gaussian process prior for the latent function with a Dirichlet process prior for the link function, extending the interpretative framework of de Finetti representation theorem and the construction of random distribution functions made by Ferguson (1973). This approach allows for flexible uncertainty modeling in both the latent score and the mapping to probabilities. We demonstrate the method performance using simulated data where it outperforms standard logistic regression.