Addressing these challenges requires scale. To that end we introduce ProteinGym, a large-scale and holistic set of benchmarks specifically designed for protein fitness prediction and design.

QA, a multiple-choice trivia dataset that is parallel in English and Basque.

In contrast, the mathematics needed to analyze such schemes is what forms the focus in Stochastic Approximation (SA) theory [2, 4]. More generally, SA refers to an iterative scheme that helps find zeroes or optimal points of a function, for which only noisy evaluationsarepossible.

Modeling the time evolution of discrete sets of items (e.g., genetic mutations) is a fundamental problem in many biomedical applications. We approach this problem through the lens of continuous-time Markov chains, and show that the resulting learning task is generally underspecified in the usual setting of cross-sectional data. We explore a perhaps surprising remedy: including a number of additional independent items can help determine time order, and hence resolve underspecifi-cation. This is in sharp contrast to the common practice of limiting the analysis to a small subset of relevant items, which is followed largely due to poor scaling of existing methods. To put our theoretical insight into practice, we develop an approximate likelihood maximization method for learning continuous-time Markov chains, which can scale to hundreds of items and is orders of magnitude faster than previous methods. We demonstrate the effectiveness of our approach on synthetic and real cancer data.

Text-to-image diffusion models have made significant progress in generating naturalistic images from textual inputs, and demonstrate the capacity to learn and represent complex visual-semantic relationships. While these diffusion models have achieved remarkable success, the underlying mechanisms driving their performance are not yet fully accounted for, with many unanswered questions surrounding what they learn, how they represent visual-semantic relationships, and why they sometimes fail to generalize.

Deep neural networks are known to be vulnerable to small adversarial perturbations in test data. To defend against adversarial attacks, probabilistic classifiers have been proposed as an alternative to deterministic ones. However, literature has conflicting findings on the effectiveness of probabilistic classifiers in comparison to deterministic ones. In this paper, we clarify the role of randomization in building adversarially robust classifiers.Given a base hypothesis set of deterministic classifiers, we show the conditions under which a randomized ensemble outperforms the hypothesis set in adversarial risk, extending previous results.Additionally, we show that for any probabilistic binary classifier (including randomized ensembles), there exists a deterministic classifier that outperforms it. Finally, we give an explicit description of the deterministic hypothesis set that contains such a deterministic classifier for many types of commonly used probabilistic classifiers, randomized ensembles and parametric/input noise injection.

Abstract--Orographic soaring can significantly extend the endurance of micro aerial vehicles (MA Vs), but circling behavior, arising from control conflicts between longitudinal and vertical axes, increases energy consumption and the risk of divergence. We propose a control switching method, named SAOS: Switched Control for Autonomous Orographic Soaring, which mitigates circling behavior by selectively controlling either the horizontal or vertical axis, effectively transforming the system from under-actuated to fully actuated during soaring. Additionally, the angle of attack is incorporated into the INDI controller to improve force estimation. Simulations with randomized initial positions and wind tunnel experiments on two MA Vs demonstrate that the SAOS improves position convergence, reduces throttle usage, and mitigates roll oscillations caused by pitch-roll coupling. These improvements enhance energy efficiency and flight stability in constrained soaring environments. The flight endurance of micro air vehicles (MA Vs) significantly constrains operational capabilities, limiting the scope of missions they can perform [1], [2]. This limitation is not solely due to inherently short flight durations, but also because take-off and landing procedures typically demand substantial time, energy, effort, and space. One potential solution to this problem lies in the advancement of battery technology, which could lead to improved efficiency. However, progress in this area has been relatively slow [3], [4]. Consequently, researchers have been exploring alternative solutions, such as using energy sources with higher energy densities or enabling mid-air refueling or recharging [5], [6]. Nevertheless, these approaches require considerable investment in hardware and system infrastructure, and often necessitate larger, heavier platforms--undermining the fundamental advantage of MA Vs being small. An alternative approach is to exploit soaring, a flight technique widely employed by birds [7]-[9] and human-piloted glider aircraft [10], [11]. Soaring takes advantage of wind energy, specifically upward vertical winds, to gain altitude or remain airborne with minimal energy expenditure. A key strength of soaring is its compatibility with existing systems: it can be integrated into any fixed-wing aircraft without requiring hardware modifications, making it a valuable complement to other endurance-enhancing strategies. V arious types of soaring techniques exist [12].

This work proposes PH-based metrics for benchmarking impedance control. A causality-consistent PH model is introduced for mass-spring-damper impedance in Cartesian space. Based on this model, a differentiable, force-torque sensing-independent, n-DoF passivity condition is derived, valid for time-varying references. An impedance fidelity metric is also defined from step-response power in free motion, capturing dynamic decoupling. The proposed metrics are validated in Gazebo simulations with a six-DoF manipulator and a quadruped leg. Results demonstrate the suitability of the PH framework for standardized impedance control benchmarking.

Language models depend on massive text corpora that are often filtered for quality, a process that can unintentionally exclude non-standard linguistic varieties, reduce model robustness and reinforce representational biases. In this paper, we argue that language models should aim to capture the full spectrum of language variation (dialectal, historical, informal, etc.) rather than relying solely on standardized text. Focusing on Basque, a morphologically rich and low-resource language, we construct new corpora combining standard, social media, and historical sources, and pre-train the BERnaT family of encoder-only models in three configurations: standard, diverse, and combined. We further propose an evaluation framework that separates Natural Language Understanding (NLU) tasks into standard and diverse subsets to assess linguistic generalization. Results show that models trained on both standard and diverse data consistently outperform those trained on standard corpora, improving performance across all task types without compromising standard benchmark accuracy. These findings highlight the importance of linguistic diversity in building inclusive, generalizable language models.