Automating repository-level software engineering tasks is a foundational challenge for autonomous code agents, largely due to the difficulty of configuring executable environments. However, manual configuration remains a labor-intensive bottleneck, necessitating a transition toward fully automated environment configuration. Existing approaches often rely on pre-defined artifacts or are restricted to specific programming languages, limiting their applicability to diverse real-world repositories. In this paper, we first propose RAT (RunAnyThing), a modular and extensible agent framework for fully automated configuration across programming languages on arbitrary repositories. RAT adopts a multi-stage pipeline that integrates language-aware abstraction, image initialization, specialized configuration toolset, and robust sandbox. Furthermore, to enable rigorous evaluation, we propose RATBench, a benchmark reflects the comprehensive coverage of real-world repositories. Extensive experiments demonstrate that RAT achieves state-of-the-art performance, improving Environment Setup Success Rate (ESSR) by an average of 36.1% over strong baselines.

As multi-agent Large Language Model (LLM) systems scale, evaluating their emergent coordination dynamics becomes increasingly critical. However, current evaluation paradigms-focused on single agents or small, explicitly structured groups-fail to capture the self-organization and viral information dynamics that arise in large, decentralized populations. We introduce a systematic evaluation framework to benchmark role specialization, information diffusion, and cooperative task resolution in open agent environments. We demonstrate this framework on the MoltBook Observatory Archive, a dataset of 2.73M interactions among 90,704 autonomous agents, establishing quantitative baselines for emergent coordination. Our evaluation reveals a pronounced core-periphery structure (silhouette 0.91), heavy-tailed cascade distributions ($α= 2.57$), and severe coordination overhead in decentralized task resolution (Cohen's $d = -0.88$ against a single-agent baseline). By providing standardized evaluation tasks and empirical baselines, our framework enables the rigorous comparison of future multi-agent protocols and establishes evaluation itself as an object of scientific study.

Despite the rapid progress of large language models (LLMs) in code generation, existing evaluations focus on functional correctness or syntactic validity, overlooking how LLMs make critical design choices such as which library or programming language to use. To fill this gap, we perform the first empirical study of LLMs' preferences for libraries and programming languages when generating code, covering eight diverse LLMs. We observe a strong tendency to overuse widely adopted libraries such as NumPy; in up to 45% of cases, this usage is not required and deviates from the ground-truth solutions. The LLMs we study also show a significant preference toward Python as their default language. For high-performance project initialisation tasks where Python is not the optimal language, it remains the dominant choice in 58% of cases, and Rust is not used once. These results highlight how LLMs prioritise familiarity and popularity over suitability and task-specific optimality; underscoring the need for targeted fine-tuning, data diversification, and evaluation benchmarks that explicitly measure language and library selection fidelity.

KAN-PCA is an autoencoder that uses a KAN as encoder and a linear map as decoder. It generalizes classical PCA by replacing linear projections with learned B-spline functions on each edge. The motivation is to capture more variance than classical PCA, which becomes inefficient during market crises when the linear assumption breaks down and correlations between assets change dramatically. We prove that if the spline activations are forced to be linear, KAN-PCA yields exactly the same results as classical PCA, establishing PCA as a special case. Experiments on 20 S&P 500 stocks (2015-2024) show that KAN-PCA achieves a reconstruction R^2 of 66.57%, compared to 62.99% for classical PCA with the same 3 factors, while matching PCA out-of-sample after correcting for data leakage in the training procedure.

We demonstrate how supervised learning can be decomposed into a two-stage procedure, where (1) all model parameters are selected in an unsupervised manner, and (2) the outputs y are added to the model, without changing the parameter values. This is achieved by a new model selection criterion that - in contrast to cross-validation - can be used also without access to y. For linear ridge regression, we bound the asymptotic out-of-sample risk of our method in terms of the optimal asymptotic risk. We also demonstrate that versions of linear and kernel ridge regression, smoothing splines, k-nearest neighbors, random forests, and neural networks, trained without access to y, perform similarly to their standard y-based counterparts. Hence, our results suggest that the difference between supervised and unsupervised learning is less fundamental than it may appear.

We consider a variant of sequential testing by betting where, at each time step, the statistician is presented with multiple data sources (arms) and obtains data by choosing one of the arms. We consider the composite global null hypothesis $\mathscr{P}$ that all arms are null in a certain sense (e.g. all dosages of a treatment are ineffective) and we are interested in rejecting $\mathscr{P}$ in favor of a composite alternative $\mathscr{Q}$ where at least one arm is non-null (e.g. there exists an effective treatment dosage). We posit an optimality desideratum that we describe informally as follows: even if several arms are non-null, we seek $e$-processes and sequential tests whose performance are as strong as the ones that have oracle knowledge about which arm generates the most evidence against $\mathscr{P}$. Formally, we generalize notions of log-optimality and expected rejection time optimality to more than one arm, obtaining matching lower and upper bounds for both. A key technical device in this optimality analysis is a modified upper-confidence-bound-like algorithm for unobservable but sufficiently "estimable" rewards. In the design of this algorithm, we derive nonasymptotic concentration inequalities for optimal wealth growth rates in the sense of Kelly [1956]. These may be of independent interest.

Average treatment effects (ATE) and conditional average treatment effects (CATE) are foundational causal estimands, but they target changes in expected outcomes and can miss treatment-induced changes in the shape of outcome distributions. A canonical failure mode occurs when control outcomes are unimodal, treated outcomes become bimodal, and both distributions have the same mean. In such cases mean-based causal estimands are zero even though the geometry and topology of the outcome law change substantially. This paper develops a topological causal framework based on persistent homology. We formalize a persistent-homology ignorability condition, define topological analogues of CATE and ATE, and prove that these estimands are identifiable up to an explicit error bound under approximate topological ignorability. We also clarify a subtle but important point: a marginal persistence-diagram effect is not identified from conditional topological ignorability alone because persistent homology does not in general commute with mixtures over covariates. To preserve the original intuition while ensuring scientific correctness, we retain the marginal effect as a motivating quantity, but place the mathematically sound conditional estimands at the center of the theory. A synthetic experiment with mean-preserving topology change shows that mean-based causal estimands remain near zero while the proposed topological effect increases sharply and remains recoverable after adjustment for confounding.

Large Language Models (LLMs) are increasingly being used to provide support and advice in personal domains such as romantic relationships, yet little is known about user perceptions of this type of advice. This study investigated how people evaluate advice on LLM-generated romantic relationships. Participants rated advice satisfaction, model reliability, and helpfulness, and completed pre- and post-measures of their general attitudes toward LLMs. Overall, the results showed participants' high satisfaction with LLM-generated advice. Greater satisfaction was, in turn, strongly and positively associated with their perceptions of the models' reliability and helpfulness. Importantly, participants' attitudes toward LLMs improved significantly after exposure to the advice, suggesting that supportive and contextually relevant advice can enhance users' trust and openness toward these AI systems.

Computation and communication in distributed LLM training and inference are traditionally optimized in isolation; expert-crafted systems such as DeepEP, FLUX, and TokenWeave show the potential of co-design but require deep systems expertise and hardware-specific tuning; CUCo is an agentic framework that automates compute-communication co-design of CUDA kernels by combining a structured design-space formalization with a correctness-first fast-path agent for reliable baselines and an evolution-driven slow-path agent for high-performance strategies, achieving up to 1.57x speedup across four multi-GPU workloads and discovering a two-stream overlap strategy on a DeepSeek-V3 MoE layer that hides dispatch behind local compute at an LLM inference cost under $10 per workload.

The rapid proliferation of high-quality synthetic data -- generated by advanced AI models or collected as auxiliary data from related tasks -- presents both opportunities and challenges for statistical inference. This paper introduces a GEneral Synthetic-Powered Inference (GESPI) framework that wraps around a broad class of statistical inference procedures to safely enhance sample efficiency by combining synthetic and real data. Our framework leverages high-quality synthetic data to boost statistical power, yet adaptively defaults to the standard method using only real data when synthetic data are of low quality. The error rate of our method remains below a user-specified bound without any distributional assumptions on the synthetic data, and decreases as the quality of the synthetic data improves. This flexibility enables seamless integration with conformal prediction, risk control, hypothesis testing, and multiple testing procedures, all without modifying the base inference method. We demonstrate the benefits of our method on challenging tasks with limited labeled data, including AlphaFold protein structure prediction, and comparing large reasoning models on complex math problems.

Cross-validation (CV) is known to provide asymptotically exact tests and confidence intervals for model improvement but only when the model comparison is relatively stable. Surprisingly, we prove that even simple, individually stable models can generate relatively unstable comparisons, calling into question the validity of CV inference. Specifically, we show that the Lasso and its close cousin, soft-thresholding, generate relatively unstable comparisons and invalid CV inferences, even in the most favorable of learning settings and when both models are individually stable. These findings highlight the importance of verifying relative stability before deploying CV for model comparison.

Embedding geometry plays a fundamental role in retrieval quality, yet dense retrievers for retrieval-augmented generation (RAG) remain largely confined to Euclidean space. However, natural language exhibits hierarchical structure from broad topics to specific entities that Euclidean embeddings fail to preserve, causing semantically distant documents to appear spuriously similar and increasing hallucination risk. To address these limitations, we introduce hyperbolic dense retrieval, developing two model variants in the Lorentz model of hyperbolic space: HyTE-FH, a fully hyperbolic transformer, and HyTE-H, a hybrid architecture projecting pre-trained Euclidean embeddings into hyperbolic space. To prevent representational collapse during sequence aggregation, we introduce the Outward Einstein Midpoint, a geometry-aware pooling operator that provably preserves hierarchical structure. On MTEB, HyTE-FH outperforms equivalent Euclidean baselines, while on RAGBench, HyTE-H achieves up to 29% gains over Euclidean baselines in context relevance and answer relevance using substantially smaller models than current state-of-the-art retrievers. Our analysis also reveals that hyperbolic representations encode document specificity through norm-based separation, with over 20% radial increase from general to specific concepts, a property absent in Euclidean embeddings, underscoring the critical role of geometric inductive bias in faithful RAG systems.

Differentially private synthetic data enables the sharing and analysis of sensitive datasets while providing rigorous privacy guarantees for individual contributors. A central challenge is to achieve strong utility guarantees for meaningful downstream analysis. Many existing methods ensure uniform accuracy over broad query classes, such as all Lipschitz functions, but this level of generality often leads to suboptimal rates for statistics of practical interest. Since many common data analysis queries exhibit smoothness beyond what worst-case Lipschitz bounds capture, we ask whether exploiting this additional structure can yield improved utility. We study the problem of generating $(\varepsilon,δ)$-differentially private synthetic data from a dataset of size $n$ supported on the hypercube $[-1,1]^d$, with utility guarantees uniformly for all smooth queries having bounded derivatives up to order $k$. We propose a polynomial-time algorithm that achieves a minimax error rate of $O_{k,d}(n^{-\min \{1, \frac{k}{d}\}})$, up to a $\log(n)$ factor. This characterization uncovers a phase transition at $k=d$. Our results generalize the Chebyshev moment matching framework of (Musco et al., 2025; Wang et al., 2016) and strictly improve the error rates for $k$-smooth queries established in \citep{wang2016differentially}. Moreover, we establish the first minimax lower bound for the utility of $(\varepsilon,δ)$-differentially private synthetic data with respect to $k$-smooth queries, extending the Wasserstein lower bound for $\varepsilon$-differential privacy in (Boedihardjo et al., 2024).

Online scams increasingly leverage fluent and context-aware social engineering strategies, creating growing demand for AI systems that explain why a message may be risky. However, explanations that cite detector-derived evidence may still semantically weaken or redirect the intended risk interpretation. We introduce VEXA: Verifying Semantic Explanation Alignment, a controlled testbed for studying the gap between lexical grounding and semantic risk alignment in AI-generated scam-risk explanations. VEXA generates ungrounded, risk-aligned, and risk-diluting explanations by independently controlling evidence grounding and semantic framing. Through LLM-as-a-judge and human evaluations, we show that explanations may continue to appear comparatively grounded even when their semantic interpretation weakens the detector's intended risk assessment. In human evaluation, risk-diluting XAI-grounded explanations retained comparatively elevated Perceived Evidence Grounding scores (3.66) despite lower Helpfulness (3.00) and Reasoning Support (3.14) scores. These findings provide controlled evidence of grounding illusion effects in AI-generated security explanations and suggest that trustworthy explanation evaluation must verify not only whether evidence is cited, but also how that evidence is interpreted.

Graph Retrieval-Augmented Generation (Graph-RAG) enhances multihop question answering by organizing corpora into knowledge graphs and routing evidence through relational structure. However, practical deployments face two persistent bottlenecks: (i) mixed-difficulty workloads where one-size-fits-all retrieval either wastes cost on easy queries or fails on hard multihop cases, and (ii) extraction loss, where graph abstraction omits fine-grained qualifiers that remain only in source text. We present A2RAG, an adaptive-and-agentic GraphRAG framework for cost-aware and reliable reasoning. A2RAG couples an adaptive controller that verifies evidence sufficiency and triggers targeted refinement only when necessary, with an agentic retriever that progressively escalates retrieval effort and maps graph signals back to provenance text to remain robust under extraction loss and incomplete graphs. Experiments on HotpotQA and 2WikiMultiHopQA demonstrate that A2RAG achieves +9.9/+11.8 absolute gains in Recall@2, while cutting token consumption and end-to-end latency by about 50% relative to iterative multihop baselines.

Accurate assessment of human epidermal growth factor receptor 2 (HER2) expression is critical for breast cancer diagnosis, prognosis, and therapy selection; yet, most existing digital HER2 scoring methods rely on bulky and expensive optical systems. Here, we present a compact and cost-effective lensfree holography platform integrated with deep learning for automated HER2 scoring of immunohistochemically stained breast tissue sections. The system captures lensfree diffraction patterns of stained HER2 tissue sections under RGB laser illumination and acquires complex field information over a sample area of ~1,250 mm^2 at an effective throughput of ~84 mm^2 per minute. To enhance diagnostic reliability, we incorporated an uncertainty quantification strategy based on Bayesian Monte Carlo dropout, which provides autonomous uncertainty estimates for each prediction and supports reliable, robust HER2 scoring, with an overall correction rate of 30.4%. Using a blinded test set of 412 unique tissue samples, our approach achieved a testing accuracy of 84.9% for 4-class (0, 1+, 2+, 3+) HER2 classification and 94.8% for binary (0/1+ vs. 2+/3+) HER2 scoring with uncertainty quantification. Overall, this lensfree holography approach provides a practical pathway toward portable, high-throughput, and cost-effective HER2 scoring, particularly suited for resource-limited settings, where traditional digital pathology infrastructure is unavailable.

Recommender systems based on Large Language Models (LLMs) are often plagued by hallucinations of out-of-domain (OOD) items. To address this, we propose RecLM, a unified framework that bridges the gap between retrieval and generation by instantiating three grounding paradigms under a single architecture: embedding-based retrieval, constrained generation over rewritten item titles, and discrete item-tokenizer generation. Using the same backbone LLM and prompts, we systematically compare these three views on public benchmarks. RecLM strictly eradicates OOD recommendations (OOD@10 = 0) across all variants, and the constrained generation variants RecLM-cgen and RecLM-token achieve overall state-of-the-art accuracy compared to both strong ID-based and LLM-based baselines. Our unified view provides a systematic basis for comparing three distinct paradigms to reduce item hallucinations, offering a practical framework to facilitate the application of LLMs to recommendation tasks. Source code is at https://github.com/microsoft/RecAI.

During 2025 and 2026, the Energy Performance of Buildings Directive is being implemented in the European Union member states, requiring all member states to have National Building Renovation Plans. In Sweden, there is no comprehensive national register of buildings with heritage values. This is seen as a barrier for the analyses underlying the development of Building Renovation Plans by the involved Swedish authorities. The purpose of this research was to assist Swedish authorities in developing information on heritage values in the Swedish building stock. Buildings in street view images from all over Sweden (N=154 710) have been analysed using multimodal Large Language Models (LLM) to assess visible aspects indicative of heritage value. Zero-shot predictions by LLMs were used as a basis for identifying buildings with potential heritage values for 5.0 million square meters of heated floor area. In this paper, the results of the predictions and lessons learned are presented and related to the development of the Swedish Building Renovation Plan as part of governance. The problems with the method and potential improvements are discussed. Risks with authorities use of LLM-based data are addressed, with a focus on issues of transparency, error detection and sycophancy.

Machine learning models for the global atmosphere that are capable of producing stable, multi-year simulations of Earth's climate have recently been developed. However, the ability of these ML models to generalize beyond the training distribution remains an open question. In this study, we evaluate the climate response of several state-of-the-art ML models (ACE2-ERA5, NeuralGCM, and cBottle) to a uniform sea surface temperature warming, a widely used benchmark for evaluating climate change. We assess each ML model's performance relative to a physics-based general circulation model (NOAA's Geophysical Fluid Dynamics Laboratory AM4) across key diagnostics, including surface air temperature, precipitation, temperature and wind profiles, and top-of-atmosphere radiation. While the ML models reproduce key aspects of the physical model response, particularly the response of precipitation, some exhibit notable departures from robust physical responses, including radiative responses and land region warming. Our results highlight the promise and current limitations of ML models for climate change applications and suggest that further improvements are needed for robust out-of-sample generalization.

Computational point-of-care (POC) sensors enable rapid, low-cost, and accessible diagnostics in emergency, remote and resource-limited areas that lack access to centralized medical facilities. These systems can utilize neural network-based algorithms to accurately infer a diagnosis from the signals generated by rapid diagnostic tests or sensors. However, neural network-based diagnostic models are subject to hallucinations and can produce erroneous predictions, posing a risk of misdiagnosis and inaccurate clinical decisions. To address this challenge, here we present an autonomous uncertainty quantification technique developed for POC diagnostics. As our testbed, we used a paper-based, computational vertical flow assay (xVFA) platform developed for rapid POC diagnosis of Lyme disease, the most prevalent tick-borne disease globally. The xVFA platform integrates a disposable paper-based assay, a handheld optical reader and a neural network-based inference algorithm, providing rapid and cost-effective Lyme disease diagnostics in under 20 min using only 20 uL of patient serum. By incorporating a Monte Carlo dropout (MCDO)-based uncertainty quantification approach into the diagnostics pipeline, we identified and excluded erroneous predictions with high uncertainty, significantly improving the sensitivity and reliability of the xVFA in an autonomous manner, without access to the ground truth diagnostic information of patients. Blinded testing using new patient samples demonstrated an increase in diagnostic sensitivity from 88.2% to 95.7%, indicating the effectiveness of MCDO-based uncertainty quantification in enhancing the robustness of neural network-driven computational POC sensing systems.

Intent-Based Networking (IBN) offers a promising paradigm for intelligent and automated network control in Industrial Internet of Things (IIoT) environments by translating high-level user intents into executable network strategies. However, frequent strategy deployment and rollback are impractical due to tightly coupled workflows and high downtime costs, while node heterogeneity and privacy constraints further complicate centralized strategy evaluation. To address these challenges, we propose a Federated Evaluation Enhanced Intent-Based Networking framework (FEIBN), which leverages large language models (LLMs) to translate user intents into structured strategy tuples and employs federated learning to support distributed strategy evaluation. To improve training efficiency and reduce communication overhead, we design a Strategy Similarity Aware Federated Learning mechanism (SSAFL), which selects nodes relevant to the task based on strategy similarity and resource status, and triggers asynchronous model uploads only when local updates are significant. Experiments demonstrate that the proposed method improves model accuracy, accelerates convergence, and reduces communication cost compared with the baselines.

Single-cell transcriptomics and proteomics have become a great source for data-driven insights into biology, enabling the use of advanced deep learning methods to understand cellular heterogeneity and gene expression at the single-cell level. With the advent of spatial-omics data, we have the promise of characterizing cells within their tissue context as it provides both spatial coordinates and intra-cellular transcriptional or protein counts. Proteomics offers a complementary view by directly measuring proteins, which are the primary effectors of cellular function and key therapeutic targets. However, existing models either ignore the spatial information or the complex genetic and proteomic programs within cells. Thus they cannot infer how cell internal regulation adapts to microenvironmental cues. Furthermore, these models often utilize fixed gene vocabularies, hindering their generalizability unseen genes. In this paper, we introduce HEIST, a hierarchical graph transformer foundation model for spatial transcriptomics and proteomics. HEIST models tissues as hierarchical graphs. The higher level graph is a spatial cell graph, and each cell in turn, is represented by its lower level gene co-expression network graph. HEIST achieves this by performing both intra-level and cross-level message passing to utilize the hierarchy in its embeddings and can thus generalize to novel datatypes including spatial proteomics without retraining. HEIST is pretrained on 22.3M cells from 124 tissues across 15 organs using spatially-aware contrastive and masked autoencoding objectives. Unsupervised analysis of HEIST embeddings reveals spatially informed subpopulations missed by prior models. Downstream evaluations demonstrate generalizability to proteomics data and state-of-the-art performance in clinical outcome prediction, cell type annotation, and gene imputation across multiple technologies.

Large language models (LLMs) have shown remarkable capabilities in code translation, yet their performance deteriorates in low-resource programming domains such as Fortran and emerging frameworks like CUDA, where high-quality parallel data are scarce. We present an automated dataset generation pipeline featuring a dual-LLM Questioner-Solver design that incorporates external knowledge from compilers and runtime feedback. Beyond traditional source-target code pair datasets, our approach additionally generates (1) verified translations with unit tests for assessing functional consistency and (2) multi-turn dialogues that capture the reasoning process behind translation refinement. Applied to Fortran-to-C++ and C++-to-CUDA, the pipeline yields 3.64k and 3.93k dialogues, respectively. Fine-tuning on this data yields dramatic improvements in functional correctness, boosting unit test success rates by over 56% on the challenging C++-to-CUDA task. We show that the generated data enables a 7B open-weight model to significantly outperform larger proprietary systems on key metrics like compilation success.

Graph spectral representations are fundamental in graph signal processing, providing a rigorous frameworkforanalyzing graph-structured data. The graph fractional Fourier transform (GFRFT) extends the graph Fourier transform (GFT) through a fractional-order parameter, enabling flexible spectral analysis with mathematical consistency. The angular graph Fourier transform (AGFT) further introduces angular control by rotating GFT eigenvectors; however, existing constructions may fail to reduce exactly to the GFT at zero angle, weakening theoretical consistency and interpretability. To address these complementary limitations, namely the lack of rotation-based basis control in GFRFT and the defective zero-angle degeneracy of AGFT, this paper proposes the rotation-parameterized graph fractional Fourier transform (RP-GFRFT), which unifies fractional order and rotation-parameterized spectral analysis. A degeneracy preserving rotation matrix family is constructed to guarantee exact GFT reduction at zero angle. TwoRP-GFRFTvariants,I-RP-GFRFTandII-RP-GFRFT,arethenformulated, with theoretical analyses confirming their unitarity, invertibility, reduction behavior, and smooth parameter dependence. The fractional order and rotation angle are jointly optimized for adaptive graph spectral filtering. Experiments on real-world signals, images, and point clouds demonstrate that RP-GFRFT improves denoising accuracy, reconstruction quality, and feature preservation over GFRFT, AGFT, and representative filtering baselines.

Attacks on machine learning models have been extensively studied through stateless optimization. In this paper, we demonstrate how a reinforcement learning (RL) agent can learn a new class of attack algorithms that generate adversarial samples. Unlike traditional adversarial machine learning (AML) methods that craft adversarial samples independently, our RL-based approach retains and exploits past attack experience to improve the effectiveness and efficiency of future attacks. We formulate adversarial sample generation as a Markov Decision Process and evaluate RL's ability to (a) learn effective and efficient attack strategies and (b) compete with state-of-the-art AML. On two image classification benchmarks, our agent increases attack success rate by up to 13.2% and decreases the average number of victim model queries per attack by up to 16.9% from the start to the end of training. In a head-to-head comparison with state-of-the-art image attacks, our approach enables an adversary to generate adversarial samples with 17% more success on unseen inputs post-training. From a security perspective, this work demonstrates a powerful new attack vector that uses RL to train agents that attack ML models efficiently and at scale.

Universality results for equivariant neural networks remain rare. Those that do exist typically hold only in restrictive settings: either they rely on regular or higher-order tensor representations, leading to impractically high-dimensional hidden spaces, or they target specialized architectures, often confined to the invariant setting. This work develops a more general account. For invariant networks, we establish a universality theorem under separation constraints, showing that the addition of a fully connected readout layer secures approximation within the class of separation-constrained continuous functions. For equivariant networks, where results are even scarcer, we demonstrate that standard separability notions are inadequate and introduce the sharper criterion of $\textit{entry-wise separability}$. We show that with sufficient depth or with the addition of appropriate readout layers, equivariant networks attain universality within the entry-wise separable regime. Together with prior results showing the failure of universality for shallow models, our findings identify depth and readout layers as a decisive mechanism for universality, additionally offering a unified perspective that subsumes and extends earlier specialized results.

Cyber Threat Intelligence (CTI) is foundational to modern cybersecurity, enabling organizations to proactively defend against evolving threats. However, the sheer volume and heterogeneity of CTI data, spanning structured knowledge bases (CVE, CWE, CAPEC, MITRE ATT&CK) and unstructured threat reports, far exceed the capacity of manual analysis. The strong contextual understanding and reasoning of Large Language Models (LLMs) have driven growing interest in applying them to CTI tasks. Yet no existing benchmark evaluates LLMs in a retrieval-augmented setting with a proper evaluation harness that grants access to the heterogeneous domain knowledge sources analysts rely on in practice. To address this gap, we present CTIConnect, a benchmark for systematically evaluating retrieval-augmented LLMs across the CTI task landscape. We construct a unified evaluation environment integrating five heterogeneous CTI sources into 1,860 expert-verified QA pairs spanning nine tasks across three categories: Entity Linking, Multi-Document Synthesis, and Entity Attribution. Extensive experiments on ten state-of-the-art LLMs reveal that the cross-source semantic gap manifests differently across task categories, demanding fundamentally different retrieval strategies, and that the performance bottleneck shifts between retrieval infrastructure and evidence utilization depending on the task. Our domain-specific strategies further outperform stronger general-purpose retrieval paradigms (retrieve-then-rerank, IRCoT), showing that closing this gap requires structural interventions rather than generic retrieval improvements. These findings hold across all ten LLMs, remain consistent on the full benchmark, and stay stable under temporal splits spanning 2008-2025. Together, they provide actionable guidance for designing scalable retrieval architectures over heterogeneous CTI ecosystems.

LLM benchmarking metrics often misstate performance and uncertainty as they rely on two assumptions that frequently do not hold in practice: (i) a sufficient number of evaluations are available for classical inference, and (ii) test prompts are independent. We propose a corrective Bayesian hierarchical model with embedding-space clustering that provides robust performance metrics in limited-data settings while correcting for prompt dependence. We apply the approach to adversarial robustness benchmarks, showing consistent recovery of clustering structure, resulting in more reliable performance metrics, with 4-73% improvements to mean absolute errors and 40-450 unit improvements to expected log posterior densities.

This work develops a computational framework that combines physics-informed neural networks with multi-patch isogeometric analysis to solve partial differential equations on complex computer-aided design geometries. The method utilizes patch-local neural networks that operate on the reference domain of isogeometric analysis. A custom output layer enables the strong imposition of Dirichlet boundary conditions. Solution conformity across interfaces between non-uniform rational B-spline patches is enforced using dedicated interface neural networks. Training is performed using the variational framework by minimizing the energy functional derived after the weak form of the partial differential equation. The effectiveness of the suggested method is demonstrated on two highly non-trivial and practically relevant use-cases, namely, a 2D magnetostatics model of a quadrupole magnet and a 3D nonlinear solid and contact mechanics model of a mechanical holder. The results show excellent agreement to reference solutions obtained with high-fidelity finite element solvers, thus highlighting the potential of the suggested neural solver to tackle complex engineering problems given the corresponding computer-aided design models.

The proliferation of open large language models (LLMs) is fostering a vibrant ecosystem in artificial intelligence (AI). However, the methods of collaboration used to develop open LLMs, both before and after their public release, have not yet been systematically studied, limiting our understanding of how open LLM projects are initiated, organised, and governed, as well as the opportunities to further foster this ecosystem. We address this gap through an exploratory analysis of open collaboration throughout the development and reuse lifecycle of open LLMs, drawing on semi-structured interviews with the developers of 14 diverse open LLM projects. These collaborations span multiple artefact domains -- including models, data, software, evaluation, compute, and community engagement -- each enabling distinct forms of participation and involving different stakeholders that evolves across the LLM development lifecycle, shifting from concentrated, selective engagement in the early stages to broader, distributed participation after model release. The open LLM developers are motivated by a variety of social, economic, and technological motivations, ranging from democratising access to AI and promoting open science to building regional ecosystems and expanding language representation. These dynamics are coordinated through a range of governance structures, typically formal and professionalised to varying degrees, including centralised company-led efforts to decentralised grassroots initiatives. We synthesise our findings in a conceptual model of open collaboration in open LLM ecosystems, provide recommendations for practice, and conclude that openness in open source AI is not a uniform property but an emergent outcome of how collaboration is organised across interconnected artefact domains, lifecycle stages, and institutional contexts.