Natural language processing (NLP) enables the understanding and generation of meaningful human language, typically using a pre-trained complex architecture on a large dataset to learn the language and next fine-tune its weights to implement a specific task. Twofold goals are examined; to understand the mechanism underlying successful pre-training and to determine the interplay between the pre-training accuracy and the fine-tuning of classification tasks. The following main results were obtained; the accuracy per token (APT) increased with its appearance frequency in the dataset, and its average over all tokens served as an order parameter to quantify pre-training success, which increased along the transformer blocks. Pre-training broke the symmetry among tokens and grouped them into finite, small, strong match token clusters, as inferred from the presented token confusion matrix. This feature was sharpened along the transformer blocks toward the output layer, enhancing its performance considerably compared with that of the embedding layer. Consequently, higher-order language structures were generated by pre-training, even though the learning cost function was directed solely at identifying a single token. These pre-training findings were reflected by the improved fine-tuning accuracy along the transformer blocks. Additionally, the output label prediction confidence was found to be independent of the average input APT, as the input meaning was preserved since the tokens are replaced primarily by strong match tokens. Finally, although pre-training is commonly absent in image classification tasks, its underlying mechanism is similar to that used in fine-tuning NLP classification tasks, hinting at its universality. The results were based on the BERT-6 architecture pre-trained on the Wikipedia dataset and fine-tuned on the FewRel and DBpedia classification tasks.

Reinforcement learning with verifiable rewards (RL VR) has emerged to be a predominant paradigm for mathematical reasoning tasks, offering stable improvements in reasoning ability. However, Outcome Reward Models (ORMs) in RL VR are too coarse-grained to distinguish flawed reasoning within correct answers or valid reasoning within incorrect answers. This lack of granularity introduces noisy and misleading gradients significantly and hinders further progress in reasoning process quality. While Process Reward Models (PRMs) offer fine-grained guidance for intermediate steps, they frequently suffer from inaccuracies and are susceptible to reward hacking. To resolve this dilemma, we introduce PRocess cOnsistency Filter (PROF), an effective data process curation method that harmonizes noisy, fine-grained process rewards with accurate, coarse-grained outcome rewards. Our approach retains correct responses with higher averaged process values and incorrect responses with lower averaged process values, while maintaining positive/negative training sample balance. Extensive experiments demonstrate that our method not only consistently improves the final accuracy over 4% compared to the blending approaches, but also strengthens the quality of intermediate reasoning steps. Codes and training recipes are available at https://github.com/Chenluye99/PROF . V erifiable rewards have spurred the widest attention recently because they can reliably improve the performance on reasoning tasks with easily verifiable outcomes, such as mathematical and coding problems (Cobbe et al., 2021; Jaech et al., 2024; Shao et al., 2024; Xiong et al., 2025b). However, since the verifiers can only verify the outcome results, the rewards are too sparse and coarse to measure and supervise the reasoning quality in intermediate steps.

Cloud platforms are increasingly relied upon to host diverse, resource-intensive workloads due to their scalability, flexibility, and cost-efficiency. In multi-tenant cloud environments, virtual machines are consolidated on shared physical servers to improve resource utilization. While virtualization guarantees resource partitioning for CPU, memory, and storage, it cannot ensure performance isolation. Competition for shared resources such as last-level cache, memory bandwidth, and network interfaces often leads to severe performance degradation. Existing management techniques, including VM scheduling and resource provisioning, require accurate performance prediction to mitigate interference. However, this remains challenging in public clouds due to the black-box nature of VMs and the highly dynamic nature of workloads. To address these limitations, we propose CloudFormer, a dual-branch Transformer-based model designed to predict VM performance degradation in black-box environments. CloudFormer jointly models temporal dynamics and system-level interactions, leveraging 206 system metrics at one-second resolution across both static and dynamic scenarios. This design enables the model to capture transient interference effects and adapt to varying workload conditions without scenario-specific tuning. Complementing the methodology, we provide a fine-grained dataset that significantly expands the temporal resolution and metric diversity compared to existing benchmarks. Experimental results demonstrate that CloudFormer consistently outperforms state-of-the-art baselines across multiple evaluation metrics, achieving robust generalization across diverse and previously unseen workloads. Notably, CloudFormer attains a mean absolute error (MAE) of just 7.8%, representing a substantial improvement in predictive accuracy and outperforming existing methods at least by 28%.

Sepsis is a serious, life-threatening condition. When treating sepsis, it is challenging to determine the correct amount of intravenous fluids and vasopressors for a given patient. While automated reinforcement learning (RL)-based methods have been used to support these decisions with promising results, previous studies have relied on relational data. Given the complexity of modern healthcare data, representing data as a graph may provide a more natural and effective approach. This study models patient data from the well-known MIMIC-III dataset as a heterogeneous graph that evolves over time. Subsequently, we explore two Graph Neural Network architectures - GraphSAGE and GATv2 - for learning patient state representations, adopting the approach of decoupling representation learning from policy learning. The encoders are trained to produce latent state representations, jointly with decoders that predict the next patient state. These representations are then used for policy learning with the dBCQ algorithm. The results of our experimental evaluation confirm the potential of a graph-based approach, while highlighting the complexity of representation learning in this domain.

We introduce the Neural Field Turing Machine (NFTM), a differentiable architecture that unifies symbolic computation, physical simulation, and perceptual inference within continuous spatial fields. NFTM combines a neural controller, continuous memory field, and movable read/write heads that perform local updates. At each timestep, the controller reads local patches, computes updates via learned rules, and writes them back while updating head positions. This design achieves linear O(N) scaling through fixed-radius neighborhoods while maintaining Turing completeness under bounded error. We demonstrate three example instantiations of NFTM: cellular automata simulation (Rule 110), physics-informed PDE solvers (2D heat equation), and iterative image refinement (CIFAR-10 inpainting). These instantiations learn local update rules that compose into global dynamics, exhibit stable long-horizon rollouts, and generalize beyond training horizons. NFTM provides a unified computational substrate bridging discrete algorithms and continuous field dynamics within a single differentiable framework.

Reasoning is a core capability in artificial intelligence systems, for which large language models (LLMs) have recently shown remarkable progress. However, most work focuses exclusively on deductive reasoning, which is problematic since other types of reasoning are also essential in solving real-world problems, and they are less explored. This work focuses on evaluating LLMs' inductive and abductive reasoning capabilities. We introduce a programmable and synthetic dataset, InAbHyD (pronounced in-a-bid), where each reasoning example consists of an incomplete world model and a set of observations. The task for the intelligent agent is to produce hypotheses to explain observations under the incomplete world model to solve each reasoning example. We propose a new metric to evaluate the quality of hypotheses based on Occam's Razor. We evaluate and analyze some state-of-the-art LLMs. Our analysis shows that LLMs can perform inductive and abductive reasoning in simple scenarios, but struggle with complex world models and producing high-quality hypotheses, even with popular reasoning-enhancing techniques such as in-context learning and RLVR.

This paper addresses the critical gap in evaluating bias in multilingual Large Language Models (LLMs), with a specific focus on Spanish language within culturally-aware Latin American contexts. Despite widespread global deployment, current evaluations remain predominantly US-English-centric, leaving potential harms in other linguistic and cultural contexts largely underexamined. We introduce a novel, culturally-grounded framework for detecting social biases in instruction-tuned LLMs. Our approach adapts the underspecified question methodology from the BBQ dataset by incorporating culturally-specific expressions and sayings that encode regional stereotypes across four social categories: gender, race, socioeconomic class, and national origin. Using more than 4,000 prompts, we propose a new metric that combines accuracy with the direction of error to effectively balance model performance and bias alignment in both ambiguous and disambiguated contexts. To our knowledge, our work presents the first systematic evaluation examining how leading commercial LLMs respond to culturally specific bias in the Spanish language, revealing varying patterns of bias manifestation across state-of-the-art models. We also contribute evidence that bias mitigation techniques optimized for English do not effectively transfer to Spanish tasks, and that bias patterns remain largely consistent across different sampling temperatures. Our modular framework offers a natural extension to new stereotypes, bias categories, or languages and cultural contexts, representing a significant step toward more equitable and culturally-aware evaluation of AI systems in the diverse linguistic environments where they operate.

--We present a system for automatic multi-axis perceptual quality prediction of generative audio, developed for Track 2 of the AudioMOS Challenge 2025. The task is to predict four Audio Aesthetic Scores--Production Quality, Production Complexity, Content Enjoyment, and Content Usefulness--for audio generated by text-to-speech (TTS), text-to-audio (TT A), and text-to-music (TTM) systems. A main challenge is the domain shift between natural training data and synthetic evaluation data. T o address this, we combine BEA Ts, a pretrained transformer-based audio representation model, with a multi-branch long short-term memory (LSTM) predictor and use a triplet loss with buffer-based sampling to structure the embedding space by perceptual similarity. Our results show that this improves embedding discriminability and generalization, enabling domain-robust audio quality assessment without synthetic training data.

Integrating Artificial Intelligence (AI) technology in electric vehicles (EV) introduces unique challenges for safety assurance, particularly within the framework of ISO 26262, which governs functional safety in the automotive domain. Traditional assessment methodologies are not geared toward evaluating AI-based functions and require evolving standards and practices. This paper explores how an independent assessment of an AI component in an EV can be achieved when combining ISO 26262 with the recently released ISO/PAS 8800, whose scope is AI safety for road vehicles. The AI-driven State of Charge (SOC) battery estimation exemplifies the process. Key features relevant to the independent assessment of this extended evaluation approach are identified. As part of the evaluation, robustness testing of the AI component is conducted using fault injection experiments, wherein perturbed sensor inputs are systematically introduced to assess the component's resilience to input variance.

Training large-scale deep learning models has become a key challenge for the scientific community and industry. While the massive use of GPUs can significantly speed up training times, this approach has a negative impact on efficiency. In this article, we present a detailed analysis of the times reported by MLPerf Training v4.1 on four workloads: BERT, Llama2 LoRA, RetinaNet, and Stable Diffusion, showing that there are configurations that optimise the relationship between performance, GPU usage, and efficiency. The results point to a break-even point that allows training times to be reduced while maximising efficiency.