arXiv:2605.24609v1 Announce Type: cross Abstract: Artificial intelligence is increasingly integrated into radiotherapy workflows, yet such pipelines remain vulnerable to out-of-distribution image data that may introduce unexpected behavior in clinical tasks. Deep learning-based anomaly detection for pelvic magnetic resonance imaging (MRI) remains largely unexplored, and transparent evaluation of its feasibility for full automation is limited. We developed and evaluated a fully automated, unsupervised anomaly-detection framework for pelvic and brain MRI. A two-stage framework was trained on reference images from public datasets: LUND-PROBE for pelvic MRI, and IXI, fastMRI, and fastMRI+ for brain MRI. In the first stage, MRI slices were compressed into discrete tokens; in the second, the distribution of normal tokens was modeled. Anomaly evidence was estimated by combining perceptual image differences with token-surprisal scores based on negative log-likelihood. Automated detection was evaluated on pelvic MRI with synthetic global and real clinical anomalies, and on brain MRI with clinically annotated fastMRI+ abnormalities. Sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and false-positive behavior in held-out normal cases were assessed. The framework achieved robust detection across hidden evaluation cohorts, with AUCs of 0.97 (95% CI, 0.95-0.98) and 0.81 (95% CI, 0.74-0.87) for pelvic and brain MRI, respectively. Heatmap analysis showed strong spatial agreement between detected anomalies and ground-truth locations, supporting localization accuracy and interpretability. These results support the potential of unsupervised anomaly detection as an automated MRI quality-control layer for radiotherapy workflows, with transparent visualization of image regions likely to compromise downstream AI-based tasks.

arXiv:2605.25101v1 Announce Type: cross Abstract: In many industrial domains, the Functional Mock-up Interface (FMI) is used to exchange simulation models as Functional Mock-up Units (FMUs) across different partners using various modelling tools. This opens up the possibilities for simulation-based verification and validation using FMUs for ensuring reliable system behaviour. However, deriving effective test oracles for these simulation models remains challenging due to the absence of explicit expected outputs. This limits the applicability of conventional testing approaches, which require access to the internal workings of the systems. Metamorphic testing (MT) addresses this limitation by leveraging metamorphic relations (MRs), but extracting such relations from specifications remains largely a manual and error-prone process. To address this challenge, we propose an LLM-powered multi-agent workflow for specification-based metamorphic testing of FMU-based simulation models. The approach takes functional and interface specifications as input and orchestrates multiple agents to extract requirements and derive MRs. These MRs are expressed using Given-When-Then patterns to structure input conditions (Given), transformations (When), and expected output behaviours (Then). These relations are then used to generate metamorphic test cases, execute simulations, and evaluate output consistency across multiple sessions. We evaluate the approach on a Lube Oil Cooling system FMU, demonstrating its ability to automatically generate meaningful MRs and corresponding test cases. Preliminary results indicate that the proposed workflow can effectively support the systematic verification and validation of dynamic simulation models by reducing manual effort and improving test generation.

arXiv:2502.08047v5 Announce Type: replace Abstract: Recent progress in GUI agents has substantially improved visual grounding, yet robust planning remains challenging, particularly when the environment deviates from a canonical initial state. In real applications, users often invoke assistance mid-workflow, where software may be partially configured, steps may have been executed in different orders, or the interface may differ from its default setup. Such task-state variability is pervasive but insufficiently evaluated in existing GUI benchmarks. To address this gap, we introduce WorldGUI, a benchmark covering ten widely used desktop and web applications with tasks instantiated under diverse, systematically constructed initial states. These variations capture realistic human-computer interaction settings and enable diagnostic evaluation of an agent's ability to recover, adapt plans, and handle non-default contexts. We further present WorldGUI-Agent, a simple and model-agnostic framework that organizes planning and execution around three critique stages, improving reliability in dynamic environments. Experiments demonstrate that state-of-the-art GUI agents exhibit substantial performance degradation under non-default initial conditions, revealing limited robustness and fragile planning behaviors. Our benchmark and framework provide a foundation for developing more adaptable and reliable GUI agents. The code and data are available at https://github.com/showlab/WorldGUI.

arXiv:2512.15922v3 Announce Type: replace Abstract: Despite initial successes and a variety of architectures, retrieval-augmented generation systems still struggle to reliably retrieve and connect the multi-step evidence required for complicated reasoning tasks. Most of the standard RAG frameworks regard all retrieved information as equally reliable, overlooking the varying credibility and interconnected nature of large textual corpora. GraphRAG approaches offer potential improvement to RAG systems by integrating knowledge graphs, which structure information into nodes and edges, capture entity relationships, and enable multi-step logical traversal. However, GraphRAG is not always an ideal solution, as it depends on high-quality graph representations of the corpus. Such representations usually rely on manually curated knowledge graphs, which are costly to construct and update, or on automated graph-construction pipelines that are often unreliable. Moreover, systems following this paradigm typically use large language models to guide graph traversal and evidence retrieval. In this paper, we propose a novel RAG framework that uses a spreading activation algorithm to retrieve information from a corpus of documents connected by an automatically constructed heterogeneous knowledge graph. This approach reduces reliance on semantic knowledge graphs, which are often incomplete due to information loss during information extraction, avoids LLM-guided graph traversal, and improves performance on multi-hop question answering. Experiments show that our method achieves better or comparable performance to several state-of-the-art RAG methods and can be integrated as a plug-and-play module with different iterative RAG pipelines. When combined with chain-of-thought iterative retrieval, it yields up to a 39% absolute improvement in answer correctness over naive RAG, while achieving these results with small open-weight language models.

arXiv:2602.07666v3 Announce Type: replace-cross Abstract: DARPA's AI Cyber Challenge (AIxCC, 2023--2025) is the largest competition to date for building fully autonomous cyber reasoning systems (CRSs) that leverage recent advances in AI -- particularly large language models (LLMs) -- to discover and remediate vulnerabilities in real-world open-source software. This paper presents the first systematic analysis of AIxCC. Drawing on design documents, source code, execution traces, and discussions with organizers and competing teams, we examine the competition's structure and key design decisions, characterize the architectural approaches of finalist CRSs, and analyze competition results beyond the final scoreboard. Our analysis reveals the factors that truly drove CRS performance, identifies genuine technical advances achieved by teams, and exposes limitations that remain open for future research. We conclude with lessons for organizing future competitions and broader insights toward deploying autonomous CRSs in practice.

Nature, Published online: 26 May 2026; doi:10.1038/s41586-026-10651-z

Author Correction: In vitro characterization of the human segmentation clock

Nature, Published online: 19 May 2026; doi:10.1038/d41586-026-01605-6

AI might jeopardize the uncertainty required in science

Multiomics profiling of healthy individuals reveals differences across molecular layers and key pathways related to immune, metabolic, and microbiome-linked processes across ethnicities, while geographic relocation reshapes these networks and influences aging trajectories.

An optogenetic model of accelerated intermittent theta burst stimulation reveals cell type-specific plasticity mechanisms and a key role for a fronto-insular circuit in driving the antidepressant effects of this treatment in humans.

Nature, Published online: 15 April 2026; doi:10.1038/s41586-026-10368-z

The E3 ligase SCFFBXO42 degrades holoenzyme-free PP2Ac in complex with the coiled-coil protein CCDC6 to maintain cancer cell fitness

Whole-brain tissue clearing and spatial transcriptomics were applied to map postnatal vascular development in the mouse brain, identifying three coordinated phases that link vascular growth to neuronal maturation.

A platform called MIST enables quantification and genotyping of infectious influenza in expelled respiratory particles, revealing diverse, individual-specific viral loads and aerosolized variants that correlate with saliva and nasopharyngeal viral loads and symptoms. Overall, the findings indicate heterogeneity in transmission potential.

Background: The World Health Organization recommends that countries routinely collect data on the behavioral and social drivers (BeSD) of vaccination to inform public health interventions that increase vaccine uptake. There is a need to identify data collection methods that can rapidly and inexpensively collect representative data, particularly in low- and middle-income countries. Objective: This study aimed to understand BeSD drivers of vaccination in the Philippines and assess the trade-offs between survey methods. We compared responses to household, mobile, and online surveys in terms of demographics, vaccination status, responses to BeSD questions, and cost. Methods: We conducted concurrent household, mobile (SMS text messaging and interactive voice response), and online surveys among caregivers of children 2 years of age and below in Regions V and XII of the Philippines, with sampling differing by survey method. We assessed, for each survey method, (1) respondent demographics (sex, age, region, and socioeconomic status) and (2) the weighted proportion of responses from caregivers of children who received at least one dose of diphtheria-pertussis-tetanus (DPT)–containing vaccine. We estimated the weighted proportion of each BeSD survey response option and calculated the financial cost (monetary outlays) per survey response from an implementer’s perspective by summing the costs incurred in each survey method and dividing by the number of responses received. Results: We surveyed a total of 1201 household respondents, 2153 mobile respondents, and 398 online respondents from January to March 2025. We found that online and mobile survey respondents were more likely to be male and have completed high school than household survey respondents. The weighted proportion of respondents indicating that their child had received at least one dose of DPT vaccine was 91.8% (n=1090; 95% CI 90%‐93.3%) for the household survey, 90.3% (n=1853) for the mobile survey, and 85% (n=346) for the online survey. With regard to vaccine demand, more than 85% of respondents in each survey method indicated that vaccines are very important, very safe, supported by family, and that they knew where to bring a child for vaccination. More than 30% of mobile and online survey respondents indicated that it was not easy to pay for vaccination. The financial cost to conduct the survey per survey response was US $2.61 for the online survey, US $6.93 for the mobile survey, and US $29.38 for the household survey. Conclusions: In the Philippines, household, mobile, and online survey methods reached caregivers of children who were unvaccinated against DPT, and these proportions were similar across survey methods. BeSD responses indicated high vaccine demand and challenges in caregivers’ cost to access vaccination. Determining the most appropriate survey method depends on trade-offs between representativeness and costs. However, areas with strong connectivity and high mobile device ownership can consider mobile and online methods as a lower-cost alternative to rapidly collect BeSD data.

Nature, Published online: 10 April 2026; doi:10.1038/s41586-026-10484-w

Author Correction: Multi-omic profiling reveals age-related immune dynamics in healthy adults

Nature, Published online: 08 April 2026; doi:10.1038/s41586-026-10168-5

The CMS experiment at CERN reports one of the highest-precision measurements of the W boson mass, finding it in line with standard model predictions and at odds with recent anomalous measurements.

Nature, Published online: 08 April 2026; doi:10.1038/d41586-026-00630-9

The latest value for the mass of a fundamental particle called the W boson is consistent with the standard model of particle physics, challenging a previous anomalous result.

arXiv:2604.03527v1 Announce Type: new Abstract: Modern agentic workflows decompose complex tasks into specialized subtasks and route them to diverse models to minimize cost without sacrificing quality. However, current routing architectures focus exclusively on performance optimization, leaving underlying trade-offs between model capability and cost unrecorded. Without clear rationale, developers cannot distinguish between intelligent efficiency -- using specialized models for appropriate tasks -- and latent failures caused by budget-driven model selection. We present Topaz, a framework that introduces formal auditability to agentic routing. Topaz replaces silent model assignments with an inherently interpretable router that incorporates three components: (i) skill-based profiling that synthesizes performance across diverse benchmarks into granular capability profiles (ii) fully traceable routing algorithms that utilize budget-based and multi-objective optimization to produce clear traces of how skill-match scores were weighed against costs, and (iii) developer-facing explanations that translate these traces into natural language, allowing users to audit system logic and iteratively tune the cost-quality tradeoff. By making routing decisions interpretable, Topaz enables users to understand, trust, and meaningfully steer routed agentic systems.

arXiv:2604.03473v1 Announce Type: cross Abstract: Uncertainty quantification (UQ) methods for large language models are predominantly designed by hand based on domain knowledge and heuristics, limiting their scalability and generality. We apply LLM-powered evolutionary search to automatically discover unsupervised UQ methods represented as Python programs. On the task of atomic claim verification, our evolved methods outperform strong manually-designed baselines, achieving up to 6.7% relative ROC-AUC improvement across 9 datasets while generalizing robustly out-of-distribution. Qualitative analysis reveals that different LLMs employ qualitatively distinct evolutionary strategies: Claude models consistently design high-feature-count linear estimators, while Gpt-oss-120B gravitates toward simpler and more interpretable positional weighting schemes. Surprisingly, only Sonnet 4.5 and Opus 4.5 reliably leverage increased method complexity to improve performance -- Opus 4.6 shows an unexpected regression relative to its predecessor. Overall, our results indicate that LLM-powered evolutionary search is a promising paradigm for automated, interpretable hallucination detector design.

arXiv:2604.04096v1 Announce Type: cross Abstract: Generative AI (GenAI) tools are increasingly integrated into software architecture research, yet the environmental impact of their computational usage remains largely undocumented. This study presents the first systematic audit of the carbon footprint of both the digital footprint from GenAI usage in research papers, and the traditional footprint from conference activities within the context of the IEEE International Conference on Software Architecture (ICSA). We report two separate carbon inventories relevant to the software architecture research community: i) an exploratory estimate of the footprint of GenAI inference usage associated with accepted papers within a research-artifact boundary, and ii) the conference attendance and operations footprint of ICSA 2025 (travel, accommodation, catering, venue energy, and materials) within the conference time boundary. These two inventories, with different system boundaries and completeness, support transparency and community reflection. We discuss implications for sustainable software architecture, including recommendations for transparency, greener conference planning, and improved energy efficiency in GenAI operations. Our work supports a more climate-conscious research culture within the ICSA community and beyond

arXiv:2604.04708v1 Announce Type: cross Abstract: High-quality bilingual resources remain a critical bottleneck for advancing multilingual NLP in low-resource settings, particularly for Bangla. To mitigate this gap, we introduce BiST, a rigorously curated Bangla-English corpus for sentence-level grammatical classification, annotated across two fundamental dimensions: syntactic structure (Simple, Complex, Compound, Complex-Compound) and tense (Present, Past, Future). The corpus is compiled from open-licensed encyclopedic sources and naturally composed conversational text, followed by systematic preprocessing and automated language identification, resulting in 30,534 sentences, including 17,465 English and 13,069 Bangla instances. Annotation quality is ensured through a multi-stage framework with three independent annotators and dimension-wise Fleiss Kappa ($\kappa$) agreement, yielding reliable and reproducible labels with $\kappa$ values of 0.82 and 0.88 for structural and temporal annotation, respectively. Statistical analyses demonstrate realistic structural and temporal distributions, while baseline evaluations show that dual-encoder architectures leveraging complementary language-specific representations consistently outperform strong multilingual encoders. Beyond benchmarking, BiST provides explicit linguistic supervision that supports grammatical modeling tasks, including controlled text generation, automated feedback generation, and cross-lingual representation learning. The corpus establishes a unified resource for bilingual grammatical modeling and facilitates linguistically grounded multilingual research.