Aiming at the problems of computational inefficiency and insufficient interpretability faced by large models in complex tasks such as multi-round reasoning and multi-modal collaboration, this study proposes a three-layer collaboration framework based on model-controller-task adaptation (MCP). By decoupling large model functions into reasoning, generation and retrieval modules, and combining reinforcement learning-driven dynamic routing algorithms and task adaptation mechanisms, the systematic integration of control theory and large model dynamic reasoning is achieved for the first time. Experiments show that the MCP framework improves the performance of cross-modal benchmarking tasks, such as GLUE, COCO, ScienceQA, etc., by 15-30% compared with the baseline model, improves the reasoning efficiency by 40%, and generates the interpretable intermediate results through the Presenter layer, obtaining 90% of the manual interpretability scores, which provides a brand-new technological path to solve the bottleneck of the practical application of the large model.

Although contrastive and other representation-learning methods have long been explored in vision and NLP, their adoption in modern time series forecasters remains limited. We believe they hold strong promise for this domain. To unlock this potential, we explicitly align past and future representations, thereby bridging the distributional gap between input histories and future targets. To this end, we introduce TimeAlign, a lightweight, plug-and-play framework that establishes a new representation paradigm, distinct from contrastive learning, by aligning auxiliary features via a simple reconstruction task and feeding them back into any base forecaster. Extensive experiments across eight benchmarks verify its superior performance. Further studies indicate that the gains arise primarily from correcting frequency mismatches between historical inputs and future outputs. Additionally, we provide two theoretical justifications for how reconstruction improves forecasting generalization and how alignment increases the mutual information between learned representations and predicted targets. The code is available at https://github.com/TROUBADOUR000/TimeAlign.

Dynamic nonlinear systems exhibit distortions arising from coupled static and dynamic effects. Their intertwined nature poses major challenges for data-driven modeling. This paper presents a theoretical framework grounded in structured decomposition, variance analysis, and task-centric complexity bounds. The framework employs a directional lower bound on interactions between measurable system components, extending orthogonality in inner product spaces to structurally asymmetric settings. This bound supports variance inequalities for decomposed systems. Key behavioral indicators are introduced along with a memory finiteness index. A rigorous power-based condition establishes a measurable link between finite memory in realizable systems and the First Law of Thermodynamics. This offers a more foundational perspective than classical bounds based on the Second Law. Building on this foundation, we formulate a `Behavioral Uncertainty Principle,' demonstrating that static and dynamic distortions cannot be minimized simultaneously. We identify that real-world systems seem to resist complete deterministic decomposition due to entangled static and dynamic effects. We also present two general-purpose theorems linking function variance to mean-squared Lipschitz continuity and learning complexity. This yields a model-agnostic, task-aware complexity metric, showing that lower-variance components are inherently easier to learn. These insights explain the empirical benefits of structured residual learning, including improved generalization, reduced parameter count, and lower training cost, as previously observed in power amplifier linearization experiments. The framework is broadly applicable and offers a scalable, theoretically grounded approach to modeling complex dynamic nonlinear systems.

Foundational modelling of multi-dimensional time-series data in industrial systems presents a central trade-off: channel-dependent (CD) models capture specific cross-variable dynamics but lack robustness and adaptability as model layers are commonly bound to the data dimensionality of the tackled use-case, while channel-independent (CI) models offer generality at the cost of modelling the explicit interactions crucial for system-level predictive regression tasks. To resolve this, we propose the Causally-Guided Pairwise Transformer (CGPT), a novel architecture that integrates a known causal graph as an inductive bias. The core of CGPT is built around a pairwise modeling paradigm, tackling the CD/CI conflict by decomposing the multidimensional data into pairs. The model uses channel-agnostic learnable layers where all parameter dimensions are independent of the number of variables. CGPT enforces a CD information flow at the pair-level and CI-like generalization across pairs. This approach disentangles complex system dynamics and results in a highly flexible architecture that ensures scalability and any-variate adaptability. We validate CGPT on a suite of synthetic and real-world industrial datasets on long-term and one-step forecasting tasks designed to simulate common industrial complexities. Results demonstrate that CGPT significantly outperforms both CI and CD baselines in predictive accuracy and shows competitive performance with end-to-end trained CD models while remaining agnostic to the problem dimensionality.

Network failure diagnosis is challenging yet critical for high-performance computing (HPC) systems. Existing methods cannot be directly applied to HPC scenarios due to data heterogeneity and lack of accuracy. This paper proposes a novel framework, called ClusterRCA, to localize culprit nodes and determine failure types by leveraging multimodal data. ClusterRCA extracts features from topologically connected network interface controller (NIC) pairs to analyze the diverse, multimodal data in HPC systems. To accurately localize culprit nodes and determine failure types, ClusterRCA combines classifier-based and graph-based approaches. A failure graph is constructed based on the output of the state classifier, and then it performs a customized random walk on the graph to localize the root cause. Experiments on datasets collected by a top-tier global HPC device vendor show ClusterRCA achieves high accuracy in diagnosing network failure for HPC systems. ClusterRCA also maintains robust performance across different application scenarios.

Generative AI, especially through large language models (LLMs), is transforming how technical knowledge can be accessed, reused, and extended. PETSc, a widely used numerical library for high-performance scientific computing, has accumulated a rich but fragmented knowledge base over its three decades of development, spanning source code, documentation, mailing lists, GitLab issues, Discord conversations, technical papers, and more. Much of this knowledge remains informal and inaccessible to users and new developers. To activate and utilize this knowledge base more effectively, the PETSc team has begun building an LLM-powered system that combines PETSc content with custom LLM tools -- including retrieval-augmented generation (RAG), reranking algorithms, and chatbots -- to assist users, support developers, and propose updates to formal documentation. This paper presents initial experiences designing and evaluating these tools, focusing on system architecture, using RAG and reranking for PETSc-specific information, evaluation methodologies for various LLMs and embedding models, and user interface design. Leveraging the Argonne Leadership Computing Facility resources, we analyze how LLM responses can enhance the development and use of numerical software, with an initial focus on scalable Krylov solvers. Our goal is to establish an extensible framework for knowledge-centered AI in scientific software, enabling scalable support, enriched documentation, and enhanced workflows for research and development. We conclude by outlining directions for expanding this system into a robust, evolving platform that advances software ecosystems to accelerate scientific discovery.

Conversational agents powered by large language models (LLMs) are rapidly becoming integral to our daily interactions, generating unprecedented amounts of conversational data. Such datasets offer a powerful lens into societal interests, trending topics, and collective concerns. Yet, existing approaches typically treat these interactions as independent and miss critical insights that could emerge from aggregating and reasoning across large-scale conversation logs. In this paper, we introduce Aggregative Question Answering, a novel task requiring models to reason explicitly over thousands of user-chatbot interactions to answer aggregative queries, such as identifying emerging concerns among specific demographics. To enable research in this direction, we construct a benchmark, WildChat-AQA, comprising 6,027 aggregative questions derived from 182,330 real-world chatbot conversations. Experiments show that existing methods either struggle to reason effectively or incur prohibitive computational costs, underscoring the need for new approaches capable of extracting collective insights from large-scale conversational data.

Recently, end-to-end (E2E) autonomous driving presents a scalable, data-driven paradigm that has garnered increasing attention [1, 2, 3]. Despite its advantages in simplifying the driving pipeline, most existing E2E approaches rely on imitation learning [4, 5] and exhibit limitations in complex, closed-loop environments. Specifically, they often suffer from causal confusion during interactive cases [6] and struggle to generalize to out-of-distribution scenarios [7]. Recent progress in mul-timodal large language models (MLLMs) [8, 9, 10] enables vision-language reasoning [11] and zero-shot generalization [12] capabilities, offering new opportunities for E2E autonomous driving. Recent efforts have explored dual-system frameworks [13, 14, 15], LLM distillation for enhancing E2E driving [16, 17], and direct trajectory prediction in textual form [18, 19, 20]. While promising, these approaches predominantly operate in open-loop settings or exhibit suboptimal performance in closed-loop evaluations. This limitation stems from their inability to perform context-aware reasoning and robust planning in closed-loop scenarios, where continuous adaptation to dynamic environments is essential [21]. We conclude three key challenges that limit the full exploitation of MLLMs'

The recent emergence of large language models (LLMs) such as GPT-3 has marked a significant paradigm shift in machine learning. Trained on massive corpora of data, these models demonstrate remarkable capabilities in language understanding, generation, summarization, and reasoning, transforming how intelligent systems process and interact with human language. Although LLMs may still seem like a recent breakthrough, the field is already witnessing the rise of a new and more general category: multi-modal, multi-task foundation models (M3T FMs). These models go beyond language and can process heterogeneous data types/modalities, such as time-series measurements, audio, imagery, tabular records, and unstructured logs, while supporting a broad range of downstream tasks spanning forecasting, classification, control, and retrieval. When combined with federated learning (FL), they give rise to M3T Federated Foundation Models (FedFMs): a highly recent and largely unexplored class of models that enable scalable, privacy-preserving model training/fine-tuning across distributed data sources. In this paper, we take one of the first steps toward introducing these models to the power systems research community by offering a bidirectional perspective: (i) M3T FedFMs for smart grids and (ii) smart grids for FedFMs. In the former, we explore how M3T FedFMs can enhance key grid functions, such as load/demand forecasting and fault detection, by learning from distributed, heterogeneous data available at the grid edge in a privacy-preserving manner. In the latter, we investigate how the constraints and structure of smart grids, spanning energy, communication, and regulatory dimensions, shape the design, training, and deployment of M3T FedFMs.

The design of the humanoid ankle is critical for safe and efficient ground interaction. Key factors such as mechanical compliance and motor mass distribution have driven the adoption of parallel mechanism architectures. However, selecting the optimal configuration depends on both actuator availability and task requirements. We propose a unified methodology for the design and evaluation of parallel ankle mechanisms. A multi-objective optimization synthesizes the mechanism geometry, the resulting solutions are evaluated using a scalar cost function that aggregates key performance metrics for cross-architecture comparison. We focus on two representative architectures: the Spherical-Prismatic-Universal (SPU) and the Revolute-Spherical-Universal (RSU). For both, we resolve the kinematics, and for the RSU, introduce a parameterization that ensures workspace feasibility and accelerates optimization. We validate our approach by redesigning the ankle of an existing humanoid robot. The optimized RSU consistently outperforms both the original serial design and a conventionally engineered RSU, reducing the cost function by up to 41% and 14%, respectively.