Modern machine learning (ML) has grown into a tightly coupled, full-stack ecosystem that combines hardware, software, network, and applications. Many users rely on cloud providers for elastic, isolated, and cost-efficient resources. Unfortunately, these platforms as a service use virtualization, which means operators have little insight into the users' workloads. This hinders resource optimizations by the operator, which is essential to ensure cost efficiency and minimize execution time. In this paper, we argue that workload knowledge is unnecessary for system-level optimization. We propose Reveal, which takes a hardware-centric approach, relying only on hardware signals - fully accessible by operators. Using low-level signals collected from the system, Reveal detects anomalies through an unsupervised learning pipeline. The pipeline is developed by analyzing over 30 popular ML models on various hardware platforms, ensuring adaptability to emerging workloads and unknown deployment patterns. Using Reveal, we successfully identified both network and system configuration issues, accelerating the DeepSeek model by 5.97%.

Pre-training has proven effective in addressing data scarcity and performance limitations in solving PDE problems with neural operators. However, challenges remain due to the heterogeneity of PDE datasets in equation types, which leads to high errors in mixed training. Additionally, dense pre-training models that scale parameters by increasing network width or depth incur significant inference costs. To tackle these challenges, we propose a novel Mixture-of-Experts Pre-training Operator Transformer (MoE-POT), a sparse-activated architecture that scales parameters efficiently while controlling inference costs. Specifically, our model adopts a layer-wise router-gating network to dynamically select 4 routed experts from 16 expert networks during inference, enabling the model to focus on equation-specific features. Meanwhile, we also integrate 2 shared experts, aiming to capture common properties of PDE and reduce redundancy among routed experts. The final output is computed as the weighted average of the results from all activated experts. We pre-train models with parameters from 30M to 0.5B on 6 public PDE datasets. Our model with 90M activated parameters achieves up to a 40% reduction in zero-shot error compared with existing models with 120M activated parameters. Additionally, we conduct interpretability analysis, showing that dataset types can be inferred from router-gating network decisions, which validates the rationality and effectiveness of the MoE architecture.

We present a transformer-based multimodal framework for generating clinically relevant captions for MRI scans. Our system combines a DEiT-Small vision transformer as an image encoder, Medi-CareBERT for caption embedding, and a custom LSTM-based decoder. The architecture is designed to semantically align image and textual embeddings, using hybrid cosine-MSE loss and contrastive inference via vector similarity. We benchmark our method on the MultiCaRe dataset, comparing performance on filtered brain-only MRIs versus general MRI images against state-of-the-art medical image captioning methods including BLIP, R2GenGPT, and recent transformer-based approaches. Results show that focusing on domain-specific data improves caption accuracy and semantic alignment. Our work proposes a scalable, interpretable solution for automated medical image reporting.

Object detection in civil engineering applications is constrained by limited annotated data in specialized domains. We introduce DINO - YOLO, a hybrid architecture combining YOLOv12 with DINOv3 self - supervised vision transformers for data - efficient detection . DINOv3 features are strategically integrated at two locations: input preprocessing (P0) and mid - backbone enhancement (P3). Experimental validation demonstrates substantial improvements: Tunnel Segment Crack detection (648 images) achieves 12.4% improveme nt, Construction PPE (1K images) gains 13.7%, and KITTI (7K images) shows 88.6% improvement, while maintaining real - time inference (30 - 47 FPS). Systematic ablation across five YOLO scales and nine DINOv3 variants reveals that Medium - scale architectures ach ieve optimal performance with DualP0P3 integration (55.77% mAP@0.5), The 2 - 4 inference overhead (21 - 33ms versus 8 - 16ms baseline) remains acceptable for field deployment on NVIDIA RTX 5090. DINO - YOLO establishes state - of - the - art performance for civil engineering datasets (<10K images) while preserving computational efficiency, providing practical solutions for construction safety monitoring and infrastructure inspection in data - constrained environments . Keywords: object detection, DINO pre - trained weights, transfer learning, YOLO, self - supervised learning, small datasets 1. I ntroduction Object detection has emerged as a fundamental computer vision task with widespread applications across numerous domains, from autonomous vehicles to industrial inspection systems. The evolution of deep learning architectures, particularly the You Only Look Once (YOLO) family of models (Khanam and Hussain, 2024; Tian et al., 2025; Wang et al., 2024; Wang and Liao, 2024; Youwai et al., 2024), has significantly advanced real - time object detection capabilities by achieving remarkable balance between accuracy and computational efficiency. However, conventional object detection frameworks face persistent challenges when deployed in specialized do mains with limited training data, where traditional random weight initialization strategies often lead to suboptimal convergence and inadequate feature representation learning.

Evaluating the reasoning ability of language models (LMs) is complicated by their extensive parametric world knowledge, where benchmark performance often reflects factual recall rather than genuine reasoning. Existing datasets and approaches (e.g., temporal filtering, paraphrasing, adversarial substitution) cannot cleanly separate the two. We present SynthWorlds, a framework that disentangles task reasoning complexity from factual knowledge. In SynthWorlds, we construct parallel corpora representing two worlds with identical interconnected structure: a real-mapped world, where models may exploit parametric knowledge, and a synthetic-mapped world, where such knowledge is meaningless. On top of these corpora, we design two mirrored tasks as case studies: multi-hop question answering and page navigation, which maintain equal reasoning difficulty across worlds. Experiments in parametric-only (e.g., closed-book QA) and knowledge-augmented (e.g., retrieval-augmented) LM settings reveal a persistent knowledge advantage gap, defined as the performance boost models gain from memorized parametric world knowledge. Knowledge acquisition and integration mechanisms reduce but do not eliminate this gap, highlighting opportunities for system improvements. Fully automatic and scalable, SynthWorlds provides a controlled environment for evaluating LMs in ways that were previously challenging, enabling precise and testable comparisons of reasoning and memorization.

We characterize how memorization is represented in transformer models and show that it can be disentangled in the weights of both language models (LMs) and vision transformers (ViTs) using a decomposition based on the loss landscape curvature. This insight is based on prior theoretical and empirical work showing that the curvature for memorized training points is much sharper than non memorized, meaning ordering weight components from high to low curvature can reveal a distinction without explicit labels. This motivates a weight editing procedure that suppresses far more recitation of untargeted memorized data more effectively than a recent unlearning method (BalancedSubnet), while maintaining lower perplexity. Since the basis of curvature has a natural interpretation for shared structure in model weights, we analyze the editing procedure extensively on its effect on downstream tasks in LMs, and find that fact retrieval and arithmetic are specifically and consistently negatively affected, even though open book fact retrieval and general logical reasoning is conserved. We posit these tasks rely heavily on specialized directions in weight space rather than general purpose mechanisms, regardless of whether those individual datapoints are memorized. We support this by showing a correspondence between task data's activation strength with low curvature components that we edit out, and the drop in task performance after the edit. Our work enhances the understanding of memorization in neural networks with practical applications towards removing it, and provides evidence for idiosyncratic, narrowly-used structures involved in solving tasks like math and fact retrieval.

Artificial intelligence (AI) has emerged as a powerful accelerator of materials discovery, yet most existing models remain problem-specific, requiring additional data collection and retraining for each new property. Here we introduce and validate GATE (Geometrically Aligned Transfer Encoder) -- a generalizable AI framework that jointly learns 34 physicochemical properties spanning thermal, electrical, mechanical, and optical domains. By aligning these properties within a shared geometric space, GATE captures cross-property correlations that reduce disjoint-property bias -- a key factor causing false positives in multi-criteria screening. To demonstrate its generalizable utility, GATE -- without any problem-specific model reconfiguration -- applied to the discovery of immersion cooling fluids for data centers, a stringent real-world challenge defined by the Open Compute Project (OCP). Screening billions of candidates, GATE identified 92,861 molecules as promising for practical deployment. Four were experimentally or literarily validated, showing strong agreement with wet-lab measurements and performance comparable to or exceeding a commercial coolant. These results establish GATE as a generalizable AI platform readily applicable across diverse materials discovery tasks.

Interpretable-by-design models are crucial for fostering trust, accountability, and safe adoption of automated decision-making models in real-world applications. In this paper we formalize the ground for the MIMOSA (Mining Interpretable Models explOiting Sophisticated Algorithms) framework, a comprehensive methodology for generating predictive models that balance interpretability with performance while embedding key ethical properties. We formally define here the supervised learning setting across diverse decision-making tasks and data types, including tabular data, time series, images, text, transactions, and trajectories. We characterize three major families of interpretable models: feature importance, rule, and instance based models. For each family, we analyze their interpretability dimensions, reasoning mechanisms, and complexity. Beyond interpretability, we formalize three critical ethical properties, namely causality, fairness, and privacy, providing formal definitions, evaluation metrics, and verification procedures for each. We then examine the inherent trade-offs between these properties and discuss how privacy requirements, fairness constraints, and causal reasoning can be embedded within interpretable pipelines. By evaluating ethical measures during model generation, this framework establishes the theoretical foundations for developing AI systems that are not only accurate and interpretable but also fair, privacy-preserving, and causally aware, i.e., trustworthy.

Semantic segmentation of electron microscopy (EM) images of biological samples remains a challenge in the life sciences. EM data captures details of biological structures, sometimes with such complexity that even human observers can find it overwhelming. We introduce ε-Seg, a method based on hierarchical variational autoencoders (HVAEs), employing center-region masking, sparse label contrastive learning (CL), a Gaussian mixture model (GMM) prior, and clustering-free label prediction. Center-region masking and the inpainting loss encourage the model to learn robust and representative embeddings to distinguish the desired classes, even if training labels are sparse (0.05% of the total image data or less). For optimal performance, we employ CL and a GMM prior to shape the latent space of the HVAE such that encoded input patches tend to cluster wrt. the semantic classes we wish to distinguish. Finally, instead of clustering latent embeddings for semantic segmentation, we propose a MLP semantic segmentation head to directly predict class labels from latent embeddings. We show empirical results of ε-Seg and baseline methods on 2 dense EM datasets of biological tissues and demonstrate the applicability of our method also on fluorescence microscopy data. Our results show that ε-Seg is capable of achieving competitive sparsely-supervised segmentation results on complex biological image data, even if only limited amounts of training labels are available.

Abstract--Large Language Models (LLMs) have shown great promise in automating data analytics tasks by interpreting natural language queries and generating multi-operation execution plans. However, existing LLM-agent-based analytics frameworks operate under the assumption of centralized data access, offering little to no privacy protection. In contrast, federated analytics (F A) enables privacy-preserving computation across distributed data sources, but lacks support for natural language input and requires structured, machine-readable queries. In this work, we present LAF A, the first system that integrates LLM-agent-based data analytics with F A. LAF A introduces a hierarchical multi-agent architecture that accepts natural language queries and transforms them into optimized, executable F A workflows. T o improve execution efficiency, an optimizer agent rewrites and merges multiple DAGs, eliminating redundant operations and minimizing computational and communicational overhead. Our experiments demonstrate that LAF A consistently outperforms baseline prompting strategies by achieving higher execution plan success rates and reducing resource-intensive F A operations by a substantial margin. This work establishes a practical foundation for privacy-preserving, LLM-driven analytics that supports natural language input in the F A setting. The rapid development of Large Language Models (LLMs) has offered unprecedented capabilities in natural language understanding, reasoning, and planning [1], significantly transforming the landscape of data analytics. LLMs can interpret complex analytical intents, generate structured code, and orchestrate multi-step tasks by interacting with external environments such as databases and computation sandboxes. These capabilities have led to the emergence of LLM-based agents that decompose high-level queries, plan analytical workflows, and execute or verify results through tool interactions.