Surveys have recently gained popularity as a tool to study large language models. By comparing models' survey responses to those of different human reference populations, researchers aim to infer the demographics, political opinions, or values best represented by current language models. In this work, we critically examine language models' survey responses on the basis of the well-established American Community Survey by the U.S. Census Bureau. Evaluating 43 different language models using de-facto standard prompting methodologies, we establish two dominant patterns. First, models' responses are governed by ordering and labeling biases, for example, towards survey responses labeled with the letter "A".

Current adversarial attacks for multi-class classifiers choose potential adversarial target classes naively based on the classifier's confidence levels. We present a novel adversarial targeting method, \textit{MALT - Mesoscopic Almost Linearity Targeting}, based on local almost linearity assumptions. Our attack wins over the current state of the art AutoAttack on the standard benchmark datasets CIFAR-100 and Imagenet and for different robust models. In particular, our attack uses a \emph{five times faster} attack strategy than AutoAttack's while successfully matching AutoAttack's successes and attacking additional samples that were previously out of reach. We additionally prove formally and demonstrate empirically that our targeting method, although inspired by linear predictors, also applies to non-linear models.

The digital exclusion of endangered languages remains a critical challenge in NLP, limiting both linguistic research and revitalization efforts. This study introduces the first computational investigation of Comanche, an Uto-Aztecan language on the verge of extinction, demonstrating how minimal-cost, community-informed NLP interventions can support language preservation. We present a manually curated dataset of 412 phrases, a synthetic data generation pipeline, and an empirical evaluation of GPT-4o and GPT-4o-mini for language identification. Our experiments reveal that while LLMs struggle with Comanche in zero-shot settings, few-shot prompting significantly improves performance, achieving near-perfect accuracy with just five examples. Our findings highlight the potential of targeted NLP methodologies in low-resource contexts and emphasize that visibility is the first step toward inclusion. By establishing a foundation for Comanche in NLP, we advocate for computational approaches that prioritize accessibility, cultural sensitivity, and community engagement.

This paper introduces marginal fairness, a new individual fairness notion for equitable decision-making in the presence of protected attributes such as gender, race, and religion. This criterion ensures that decisions--based on generalized distortion risk measures--are insensitive to distributional perturbations in protected attributes, regardless of whether these attributes are continuous, discrete, categorical, univariate, or multivariate. To operationalize this notion and reflect real-world regulatory environments (such as the EU gender-neutral pricing regulation), we model business decision-making in highly regulated industries (such as insurance and finance) as a two-step process: (i) a predictive modeling stage, in which a prediction function for the target variable (e.g., insurance losses) is estimated based on both protected and non-protected covariates; and (ii) a decision-making stage, in which a generalized distortion risk measure is applied to the target variable, conditional only on non-protected covariates, to determine the decision. In this second step we modify the risk measure such that the decision becomes insensitive to the protected attribute, thus enforcing fairness to ensure equitable outcomes under risk-sensitive, regulatory constraints. Furthermore, by utilising the concept of cascade sensitivity, we extend the marginal fairness framework to capture how dependencies between covariates propagate the influence of protected attributes through the modeling pipeline. A numerical study and an empirical implementation using an auto insurance dataset demonstrate how the framework can be applied in practice.

Fine-tuning large language models (LLMs) is intended to improve their reasoning capabilities, yet we uncover a counterintuitive effect: models often forget how to solve problems they previously answered correctly during training. We term this phenomenon T emporal F orgettingand show that it is widespread across model sizes, fine-tuning methods (both Reinforcement Learning and Supervised Fine-Tuning), and multiple reasoning benchmarks. Our analysis reveals that 6.4% to 56.1% of final errors were once solved correctly at an earlier checkpoint. Inspired by the phenomenon of Temporal Forgetting, we proposed T emporal Sampling, a simple decoding strategy that draws outputs from multiple checkpoints along the training trajectory. This approach recovers forgotten solutions without retraining or ensembling, and leads to significant improvements in reasoning performance, gains from 4 to 19 points in Pass@ k and consistent gains for majority-voting and Best-of-N across several benchmarks. To make Temporal Sampling deployment-friendly, we extend it to LoRA-adapted models. By leveraging the temporal diversity inherent in training, Temporal Sampling offers a practical, compute-efficient way to surface hidden reasoning ability and rethink how we evaluate LLMs.Figure 1: (a) We observed that during RL training process of Deepseek-R1-1.5B model, 76.7% of AIME problems were solved correctly at some intermediate checkpoint, yet only 30% remained correct in the final model.

Artificial Intelligence (AI) is one of the major technological advancements of this century, bearing incredible potential for users through AI-powered applications and tools in numerous domains. Being often black-box (i.e., its decision-making process is unintelligible), developers typically resort to eXplainable Artificial Intelligence (XAI) techniques to interpret the behaviour of AI models to produce systems that are transparent, fair, reliable, and trustworthy. However, presenting explanations to the user is not trivial and is often left as a secondary aspect of the system's design process, leading to AI systems that are not useful to end-users. This paper presents a Systematic Literature Review on Explanation User Interfaces (XUIs) to gain a deeper understanding of the solutions and design guidelines employed in the academic literature to effectively present explanations to users. To improve the contribution and real-world impact of this survey, we also present a framework for Human-cEnteRed developMent of Explainable user interfaceS (HERMES) to guide practitioners and academics in the design and evaluation of XUIs.

Multi-Agent Reinforcement Learning (MARL) has shown great potential as an adaptive solution for addressing modern cybersecurity challenges. MARL enables decentralized, adaptive, and collaborative defense strategies and provides an automated mechanism to combat dynamic, coordinated, and sophisticated threats. This survey investigates the current state of research in MARL applications for automated cyber defense (ACD), focusing on intruder detection and lateral movement containment. Additionally, it examines the role of Autonomous Intelligent Cyber-defense Agents (AICA) and Cyber Gyms in training and validating MARL agents. Finally, the paper outlines existing challenges, such as scalability and adversarial robustness, and proposes future research directions. This also discusses how MARL integrates in AICA to provide adaptive, scalable, and dynamic solutions to counter the increasingly sophisticated landscape of cyber threats. It highlights the transformative potential of MARL in areas like intrusion detection and lateral movement containment, and underscores the value of Cyber Gyms for training and validation of AICA.

Diffusion models have emerged as powerful generative models for graph generation, yet their use for conditional graph generation remains a fundamental challenge. In particular, guiding diffusion models on graphs under arbitrary reward signals is difficult: gradient-based methods, while powerful, are often unsuitable due to the discrete and combinatorial nature of graphs, and non-differentiable rewards further complicate gradient-based guidance. We propose Graph Guided Diffusion (GGDiff), a novel guidance framework that interprets conditional diffusion on graphs as a stochastic control problem to address this challenge. GGDiff unifies multiple guidance strategies, including gradient-based guidance (for differentiable rewards), control-based guidance (using control signals from forward reward evaluations), and zero-order approximations (bridging gradient-based and gradient-free optimization). This comprehensive, plug-and-play framework enables zero-shot guidance of pre-trained diffusion models under both differentiable and non-differentiable reward functions, adapting well-established guidance techniques to graph generation--a direction largely unexplored. Our formulation balances computational efficiency, reward alignment, and sample quality, enabling practical conditional generation across diverse reward types. We demonstrate the efficacy of GGDiff in various tasks, including constraints on graph motifs, fairness, and link prediction, achieving superior alignment with target rewards while maintaining diversity and fidelity.

--Machine translation has become a critical tool in bridging linguistic gaps, especially between languages as diverse as English and Hindi. This paper comprehensively evaluates various machine translation models for translating between English and Hindi. We assess the performance of these models using a diverse set of automatic evaluation metrics, both lexical and machine learning-based metrics. The study aims to provide insights into the effectiveness of different machine translation approaches in handling both general and specialized language domains. Results indicate varying performance levels across different metrics, highlighting strengths and areas for improvement in current translation systems.

This paper presents the design, implementation, and evaluation of an IoT-based robotic system for mapping and monitoring indoor air quality. The primary objective was to develop a mobile robot capable of autonomously mapping a closed environment, detecting concentrations of CO$_2$, volatile organic compounds (VOCs), smoke, temperature, and humidity, and transmitting real-time data to a web interface. The system integrates a set of sensors (SGP30, MQ-2, DHT11, VL53L0X, MPU6050) with an ESP32 microcontroller. It employs a mapping algorithm for spatial data acquisition and utilizes a Mamdani fuzzy logic system for air quality classification. Empirical tests in a model room demonstrated average localization errors below $5\%$, actuator motion errors under $2\%$, and sensor measurement errors within $12\%$ across all modalities. The contributions of this work include: (1) a low-cost, integrated IoT robotic platform for simultaneous mapping and air quality detection; (2) a web-based user interface for real-time visualization and control; and (3) validation of system accuracy under laboratory conditions.