Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles

The proliferation of large language models (LLMs) and modular skills has endowed autonomous agents with increasingly powerful capabilities. Existing frameworks typically rely on monolithic LLMs and fixed logic to interface with these skills. This gives rise to a critical bottleneck: different LLMs offer distinct advantages across diverse domains, yet current frameworks fail to exploit the complementary strengths of models and skills, thereby limiting their performance on downstream tasks. In this paper, we present Maestro (Multimodal Agent for Expert-Skill Targeted Reinforced Orchestration), a Reinforcement Learning (RL)-driven orchestration framework that reframes heterogeneous multimodal tasks as a sequential decision-making process over a hierarchical model-skill registry. Rather than consolidating all knowledge into a single model, Maestro trains a lightweight policy to dynamically compose ensembles of frozen expert models and a two-tier skill library, deciding at each step whether to invoke an external expert, which model-skill pair to select, and when to terminate. The policy is optimized via outcome-based RL, requiring no step-level supervision. We evaluate Maestro across ten representative multimodal benchmarks spanning mathematical reasoning, chart understanding, high-resolution perception, and domain-specific analysis. With only a 4B orchestrator, Maestro achieves an average accuracy of 70.1%, surpassing both GPT-5 (69.3%) and Gemini-2.5-Pro (68.7%). Crucially, the learned coordination policy generalizes to unseen models and skills without retraining: augmenting the registry with out-of-domain experts yields a 59.5% average on four challenging benchmarks, outperforming all closed-source baselines. Maestro further maintains high computational efficiency with low latency. The source code is available at https://github.com/jinyangwu/Maestro.

CASE-NET: Deep Spatio-Temporal Representation Learning via Causal Attention and Channel Recalibration for Multivariate Time Series Classification

Multivariate time series (MTS) classification is foundational to pervasive computing and financial analysis, yet existing multi-scale paradigms are often constrained by suboptimal representation fidelity. We identify two critical bottlenecks: temporal non-causality in standard encoders that induces temporal confounding in non-stationary dynamics, and the absence of explicit channel saliency mechanisms that allows noise to contaminate the latent space. To address these challenges, we propose the Causal Attention and Spatio-temporal Encoder Network (CASE-NET), an architecture designed for structural manifold pre-conditioning. CASE-NET synergizes a Causal Temporal Encoder, which enforces physical arrow-of-time constraints via masked self-attention and causal convolutions, with an Adaptive Channel Recalibration module functioning as an information bottleneck to suppress detrimental noise. Comprehensive evaluations across six heterogeneous domains demonstrate that CASE-NET establishes new state-of-the-art benchmarks on four tasks, achieving a peak accuracy of 98.6% on the AWR dataset and superior robustness in non-stationary regimes.

Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction

Recent reviews find that the vast majority of published healthcare federated learning (FL) studies never reach real-world deployment. We developed an embedding-based FL pipeline for iron deficiency prediction from routine full blood count (FBC) data and deployed it across real institutional environments at Amsterdam University Medical Centre (AUMC) and NHS Blood and Transplant (NHSBT), two clinical environments that differ markedly in iron deficiency prevalence, ferritin distribution, and subject populations. A frozen domain-specific haematology foundation model, DeepCBC, performs site-local representation extraction, restricting federated training to a compact downstream classifier and substantially reducing recurrent communication relative to full-encoder federation. The two clinical datasets are structurally not independent and identically distributed (non-IID), with heterogeneity arising from distinct population differences rather than sampling artefacts. Runtime governance is enforced by FLA$^3$, a healthcare-oriented FL platform providing study-scoped execution, policy-based authorisation, and signed audit logging. Standard sample-size-weighted aggregation (FedAvg) reduced the area under the receiver operating characteristic curve (ROC-AUC) at both sites relative to local-only training, as the global update was biased towards the larger AUMC distribution. FedMAP, a personalised aggregation method, raised ROC-AUC from 0.9470 to 0.9594 at AUMC and from 0.8558 to 0.8671 at NHSBT relative to local-only training, achieving the highest macro ROC-AUC of 0.9133 and the best macro balanced accuracy overall. These results support personalised aggregation in clinical federations where client sample size and task relevance diverge substantially.

GeoDiff-SAR II: 3D-Driven Foundation Diffusion Models for SAR Generation via Decoupled Control

Existing Synthetic Aperture Radar (SAR) image generation methods still lack reliable controllability over key imaging parameters, particularly azimuth angle, depression angle, and polarization mode. Our preliminary GeoDiff-SAR supported limited azimuth completion, but remained ineffective for large missing azimuth sectors and did not provide unified control over multiple imaging conditions. To address this problem, we propose GeoDiff-SAR II, a 3D model-guided decoupled framework for controllable SAR image generation. The proposed framework imposes controllability through physically grounded geometric-electromagnetic cues rather than image intensity alone. We introduce a Geometric-Electromagnetic Conditioning Map (GECM), a structured intermediate representation that encodes the target pose map and dominant scattering centers, thereby decoupling macroscopic geometry from microscopic scattering responses. During training, GECMs are derived from real sparse-azimuth SAR images. During inference, the same representation is rendered directly from a 3D CAD model under specified azimuth, depression angle, and polarization conditions, enabling physically consistent control across large viewpoint gaps. The imaging parameters are further converted into text conditions, while the GECM is injected through ControlNet to provide explicit spatial guidance. Combined with Low-Rank Adaptation (LoRA) on a FLUX backbone, the proposed framework unifies geometric-electromagnetic conditioning and parameter-aware generation within a single process. Experiments on simulated and real datasets demonstrate controllable generation over key SAR imaging parameters, stable generalization across large azimuth gaps, and consistent improvements in image fidelity, physical consistency, and downstream Automatic Target Recognition (ATR) performance.

DAFT-s-AFDM Enabled ISAC Systems: Ambiguity Function Analysis and Waveform Design

Discrete affine Fourier transform spread affine frequency division multiplexing (DAFT-s-AFDM) is a promising waveform for integrated sensing and communication (ISAC) due to its low peak-to-average power ratio, robustness to Doppler shifts, and reduced multiuser interference in the uplink transmission. This paper presents a comprehensive ambiguity function (AF) analysis of DAFT-s-AFDM and derives the closed-form expression for the AF magnitude expectation. Several key insights into the impact of DAFT-s-AFDM parameters on ISAC performance are revealed, thus providing concrete guidance for the subsequent waveform design. Building on these insights, a novel probabilistic constellation shaping (PCS) framework is proposed for ISAC waveform enhancement, where the communication throughput and the sensing AF characteristics are jointly optimized by addressing a multi-objective problem. An efficient algorithm based on a closed-form bit error rate expression is developed to obtain the Pareto-optimal solutions. Extensive simulations validate the theoretical results and that the proposed PCS-enhanced DAFT-s-AFDM can significantly outperform the classical counterparts, achieving a superior and highly controllable tradeoff between the dual-functional performances.

Evaluating the Utility of Personal Health Records in Personalized Health AI

Patient-managed Personal Health Records (PHRs) promises to empower patients to better understand their health; but information in the record is complex, potentially hindering insights. In this study, we assess the potential of large language models (LLMs, Gemini 3.0 Flash) to provide helpful answers to user health queries, when provided clinical data from PHRs as context. A total of 2,257 user queries were drawn from 3 different distributions to represent patient questions: shorter web search queries, longer questions derived from templates of chatbot conversations, and questions patients asked to their healthcare team (patient calls). Queries were matched with de-identified PHRs (from a pool of 1,945). Gemini responses were generated (1) without PHR context; (2) with a basic summary of demographics, conditions, and medications; (3) with full, extensive clinical notes. For evaluation, we leveraged an existing rating framework (SHARP), and developed a new framework for specific error modes when interpreting PHRs. Evaluation was performed using autoraters for the full set, and with clinician ratings for a subset (n=95), with both sets of raters knowing the full PHR context. We see significant improvements in the helpfulness of answers to all question types with PHR data (p < 0.001, paired t-test). We also observe potential gains in safety, accuracy, relevance and personalization of answers. Our PHR evaluation framework further identifies gaps in LLM understanding of particular aspects of complex PHRs, such as temporal disorientation, and rare but meaningful confabulations. These results suggest potential for PHR data to help people with a wide range of user needs; and provide a framework for monitoring for gaps in LLM answers based on PHR context. This study motivates further work to assess and realize potential benefits to users from understanding their health records.

Herculean: An Agentic Benchmark for Financial Intelligence

As AI agents improve, the central question is no longer whether they can solve isolated well-defined financial tasks, but whether they can reliably carry out financial professional work. Existing financial benchmarks offer only a partial view of this ability, as they primarily evaluate static competencies such as question answering, retrieval, summarization, and classification. We introduce Herculean, the first skilled benchmark for agentic financial intelligence spanning four representative workflows, including Trading, Hedging, Market Insights, and Auditing. Each workflow is instantiated as a standardized MCP-based skill environment with its own tools, interaction dynamics, constraints, and success criteria, enabling consistent end-to-end assessment of heterogeneous agent systems. Across frontier agents, we find agents perform relatively well on Trading and Market Insights, but struggle substantially on Hedging and Auditing, where long-horizon coordination, state consistency, and structured verification are critical. Overall, our results point to a key gap in current agents in turning financial reasoning into dependable workflow execution in high-stakes financial workflows.

The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations

Large language models confidently produce outdated answers, and no existing method can detect them. We show this is not an engineering failure but a structural one: temporal drift, whether a stored fact has changed since training, is encoded as a direction in the residual stream geometrically orthogonal to both correctness and uncertainty. Any method operating on correctness or uncertainty signals is therefore blind to drift by construction. We verify this across six instruction-tuned models. A linear probe trained directly on drift labels achieves AUROC $0.83$--$0.95$; methods based on token entropy, semantic entropy, CCS, and SAPLMA all remain near chance ($0.49$--$0.57$). Five tests confirm the geometric orthogonality: weight cosines ($|\cos| \leq 0.14$), score correlations ($|r| \leq 0.20$), bidirectional null-space projection ($|Δ| \leq 0.008$), iterative null-space projection with $k{=}10$, and difference-of-means dissociation. Mechanistically, the MLP retrieval circuit produces identical dynamics for stale recall and confabulation ($r > 0.81$, six models), explaining why output confidence cannot separate them. A cross-cutoff experiment holds inputs constant and varies only the model: the probe fires on the model whose training predates the fact's transition and stays silent otherwise ($P(A{>}B) = 0.975$--$0.998$, twelve model pairs), confirming it reads model-internal knowledge state rather than input properties. Our code and datasets will be publicly released.

RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI

Embodied AI is a prominent research topic in both academia and industry. Current research centers on completing tasks based on explicit user instructions. However, for robots to integrate into human society, they must understand which actions are permissible and which are prohibited, even without explicit commands. We refer to the user-guided AI as passive intelligence and the unguided AI as active intelligence. This paper introduces RobotEQ, the first benchmark for active intelligence, aiming to assess whether existing models can comprehend and adhere to social norms in embodied scenarios. First, we construct RobotEQ-Data, a dataset consisting of 1,900 egocentric images, spanning 10 representative embodied categories and 56 subcategories. Through extensive manual annotation, we provide 5,353 action judgment questions and 1,286 spatial grounding questions, specifying appropriate robot actions across diverse scenarios. Furthermore, we establish RobotEQ-Bench to evaluate the performance of state-of-the-art models on this task. Experimental results show that current models still fall short in achieving reliable active intelligence, particularly in spatial grounding. Meanwhile, we observe that leveraging RAG techniques to incorporate external social norm knowledge bases can generally enhance performance. This work can facilitate the transition of robotics from user-guided passive manipulation to active social compliance.

Posterior-First Neural PDE Simulation: Inferring Hidden Problem State from a Single Field

Neural PDE simulators often receive only a single observed field at deployment. In this setting, a field-to-future predictor can collapse distinct latent problem states into the same deterministic interface, losing the ambiguity needed for reliable rollout and downstream decisions. We propose posterior-first neural PDE simulation: first infer a posterior over the minimal task-sufficient problem state, then condition prediction on that posterior. The resulting theory connects the object, the learning target, and the failure mode: Bayes downstream values factor through this posterior, refinement labels make it learnable by proper scoring rules, and deterministic collapse incurs an ambiguity barrier whenever the true posterior is non-Dirac. Synthetic exact-ambiguity experiments show that point-versus-posterior gaps track the predicted barrier. On metadata-hidden PDEBench tasks, posterior recovery reduces pooled rollout nRMSE from 0.175 to 0.132, closing 59.4% of the direct-to-oracle gap. These results suggest that single-observation neural PDE simulation should be posterior-first rather than monolithic field-to-future prediction.