Solving time-dependent partial differential equations (PDEs) is fundamental to modeling critical phenomena across science and engineering. Physics-Informed Neural Networks (PINNs) solve PDEs using deep learning. However, PINNs perform pointwise predictions that neglect the autoregressive property of dynamical systems, leading to instabilities and inaccurate predictions. We introduce Physics-Informed Autoregressive Networks (PIANO) -- a framework that redesigns PINNs to model dynamical systems. PIANO operates autoregressively, explicitly conditioning future predictions on the past. It is trained through a self-supervised rollout mechanism while enforcing physical constraints. We present a rigorous theoretical analysis demonstrating that PINNs suffer from temporal instability, while PIANO achieves stability through autoregressive modeling. Extensive experiments on challenging time-dependent PDEs demonstrate that PIANO achieves state-of-the-art performance, significantly improving accuracy and stability over existing methods. We further show that PIANO outperforms existing methods in weather forecasting.

Multi-modal creative writing (MMCW) aims to produce illustrated articles. Unlike common multi-modal generative (MMG) tasks such as storytelling or caption generation, MMCW is an entirely new and more abstract challenge where textual and visual contexts are not strictly related to each other. Existing methods for related tasks can be forcibly migrated to this track, but they require specific modality inputs or costly training, and often suffer from semantic inconsistencies between modalities. Therefore, the main challenge lies in economically performing MMCW with flexible interactive patterns, where the semantics between the modalities of the output are more aligned. In this work, we propose FlexMUSE with a T2I module to enable optional visual input. FlexMUSE promotes creativity and emphasizes the unification between modalities by proposing the modality semantic alignment gating (msaGate) to restrict the textual input. Besides, an attention-based cross-modality fusion is proposed to augment the input features for semantic enhancement. The modality semantic creative direct preference optimization (mscDPO) within FlexMUSE is designed by extending the rejected samples to facilitate the writing creativity. Moreover, to advance the MMCW, we expose a dataset called ArtMUSE which contains with around 3k calibrated text-image pairs. FlexMUSE achieves promising results, demonstrating its consistency, creativity and coherence.

Visual Foundation Models (VFMs) are becoming ubiquitous in computer vision, powering systems for diverse tasks such as object detection, image classification, segmentation, pose estimation, and motion tracking. VFMs are capitalizing on seminal innovations in deep learning models, such as LeNet-5, AlexNet, ResNet, VGGNet, InceptionNet, DenseNet, YOLO, and ViT, to deliver superior performance across a range of critical computer vision applications. These include security-sensitive domains like biometric verification, autonomous vehicle perception, and medical image analysis, where robustness is essential to fostering trust between technology and the end-users. This article investigates network robustness requirements crucial in computer vision systems to adapt effectively to dynamic environments influenced by factors such as lighting, weather conditions, and sensor characteristics. We examine the prevalent empirical defenses and robust training employed to enhance vision network robustness against real-world challenges such as distributional shifts, noisy and spatially distorted inputs, and adversarial attacks. Subsequently, we provide a comprehensive analysis of the challenges associated with these defense mechanisms, including network properties and components to guide ablation studies and benchmarking metrics to evaluate network robustness.

With the rapid evolution of technology and the Internet, the proliferation of fake news on social media has become a critical issue, leading to widespread misinformation that can cause societal harm. Traditional fact checking methods are often too slow to prevent the dissemination of false information. Therefore, the need for rapid, automated detection of fake news is paramount. We introduce DaCFake, a novel fake news detection model using a divide and conquer strategy that combines content and context based features. Our approach extracts over eighty linguistic features from news articles and integrates them with either a continuous bag of words or a skipgram model for enhanced detection accuracy. We evaluated the performance of DaCFake on three datasets including Kaggle, McIntire + PolitiFact, and Reuter achieving impressive accuracy rates of 97.88%, 96.05%, and 97.32%, respectively. Additionally, we employed a ten-fold cross validation to further enhance the model's robustness and accuracy. These results highlight the effectiveness of DaCFake in early detection of fake news, offering a promising solution to curb misinformation on social media platforms.

Model merging is a powerful technique for integrating the specialized knowledge of multiple machine learning models into a single model. However, existing methods require manually partitioning model parameters into fixed groups for merging, which restricts the exploration of potential combinations and limits performance. To overcome these limitations, we propose Model Merging of Natural Niches (M2N2), an evolutionary algorithm with three key features: (1) dynamic adjustment of merging boundaries to progressively explore a broader range of parameter combinations; (2) a diversity preservation mechanism inspired by the competition for resources in nature, to maintain a population of diverse, high-performing models that are particularly well-suited for merging; and (3) a heuristicbased attraction metric to identify the most promising pairs of models for fusion. Our experimental results demonstrate, for the first time, that model merging can be used to evolve models entirely from scratch. Specifically, we apply M2N2 to evolve MNIST classifiers from scratch and achieve performance comparable to CMA-ES, while being computationally more efficient. Furthermore, M2N2 scales to merge specialized language and image generation models, achieving state-of-the-art performance. Notably, it preserves crucial model capabilities beyond those explicitly optimized by the fitness function, highlighting its robustness and versatility. Our code is available at https://github.com/SakanaAI/natural_niches

Flow-guided Localization (FGL) enables the identification of spatial regions within the human body that contain an event of diagnostic interest. FGL does that by leveraging the passive movement of energy-constrained nanodevices circulating through the bloodstream. Existing FGL solutions rely on graph models with fixed topologies or handcrafted features, which limit their adaptability to anatomical variability and hinder scalability. In this work, we explore the use of Set Transformer architectures to address these limitations. Our formulation treats nanodevices' circulation time reports as unordered sets, enabling permutation-invariant, variable-length input processing without relying on spatial priors. To improve robustness under data scarcity and class imbalance, we integrate synthetic data generation via deep generative models, including CGAN, WGAN, WGAN-GP, and CVAE. These models are trained to replicate realistic circulation time distributions conditioned on vascular region labels, and are used to augment the training data. Our results show that the Set Transformer achieves comparable classification accuracy compared to Graph Neural Networks (GNN) baselines, while simultaneously providing by-design improved generalization to anatomical variability. The findings highlight the potential of permutation-invariant models and synthetic augmentation for robust and scalable nanoscale localization.

Cross-organizational collaboration in Model-Based Systems Engineering (MBSE) faces many challenges in achieving semantic alignment across independently developed system models. SysML v2 introduces enhanced structural modularity and formal semantics, offering a stronger foundation for interoperable modeling. Meanwhile, GPT-based Large Language Models (LLMs) provide new capabilities for assisting model understanding and integration. This paper proposes a structured, prompt-driven approach for LLM-assisted semantic alignment of SysML v2 models. The core contribution lies in the iterative development of an alignment approach and interaction prompts, incorporating model extraction, semantic matching, and verification. The approach leverages SysML v2 constructs such as alias, import, and metadata extensions to support traceable, soft alignment integration. It is demonstrated with a GPT-based LLM through an example of a measurement system. Benefits and limitations are discussed.

The brain-computer interface (BCI) establishes a non-muscle channel that enables direct communication between the human body and an external device. Electroencephalography (EEG) is a popular non-invasive technique for recording brain signals. It is critical to process and comprehend the hidden patterns linked to a specific cognitive or motor task, for instance, measured through the motor imagery brain-computer interface (MI-BCI). A significant challenge is presented by classifying motor imagery-based electroencephalogram (MI-EEG) tasks, given that EEG signals exhibit nonstationarity, time-variance, and individual diversity. Obtaining good classification accuracy is also very difficult due to the growing number of classes and the natural variability among individuals. To overcome these issues, this paper proposes a novel method for classifying EEG motor imagery signals that extracts features efficiently with Minimally Random Convolutional Kernel Transform (MiniRocket), a linear classifier then uses the extracted features for activity recognition. Furthermore, a novel deep learning based on Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) architecture to serve as a baseline was proposed and demonstrated that classification via MiniRocket's features achieves higher performance than the best deep learning models at lower computational cost. The PhysioNet dataset was used to evaluate the performance of the proposed approaches. The proposed models achieved mean accuracy values of 98.63% and 98.06% for the MiniRocket and CNN-LSTM, respectively. The findings demonstrate that the proposed approach can significantly enhance motor imagery EEG accuracy and provide new insights into the feature extraction and classification of MI-EEG.

Usability describes a set of essential quality attributes of user interfaces (UI) that influence human-computer interaction. Common evaluation methods, such as usability testing and inspection, are effective but resource-intensive and require expert involvement. This makes them less accessible for smaller organizations. Recent advances in multimodal LLMs offer promising opportunities to automate usability evaluation processes partly by analyzing textual, visual, and structural aspects of software interfaces. To investigate this possibility, we formulate usability evaluation as a recommendation task, where multimodal LLMs rank usability issues by severity. We conducted an initial proof-of-concept study to compare LLM-generated usability improvement recommendations with usability expert assessments. Our findings indicate the potential of LLMs to enable faster and more cost-effective usability evaluation, which makes it a practical alternative in contexts with limited expert resources.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable multimodal understanding capabilities in Visual Question Answering (VQA) tasks by integrating visual and textual features. However, under the challenging ten-choice question evaluation paradigm, existing methods still exhibit significant limitations when processing PDF documents with complex layouts and lengthy content. Notably, current mainstream models suffer from a strong bias toward English training data, resulting in suboptimal performance for Japanese and other language scenarios. To address these challenges, this paper proposes a novel Japanese PDF document understanding framework that combines multimodal hierarchical reasoning mechanisms with Colqwen-optimized retrieval methods, while innovatively introducing a semantic verification strategy through sub-question decomposition. Experimental results demonstrate that our framework not only significantly enhances the model's deep semantic parsing capability for complex documents, but also exhibits superior robustness in practical application scenarios.