arXiv:2606.30444v2 Announce Type: replace-cross Abstract: Neural networks are known to be susceptible to over-reliance on spurious correlations. However, the precise mechanism by which models exploit shortcut features is not fully understood, and algorithms to mitigate this behavior rely on as yet unjustified assumptions about the learned representations. In this work, we provide the first end-to-end theoretical characterization of spurious feature learning for two-layer ReLU neural networks trained by online minibatch SGD on the logistic loss. We consider data drawn from the high-dimensional Boolean hypercube with a quadratic signal function (namely XOR) and a linear spurious correlation. We show that SGD learns the spurious feature first, and exponentially fast. Moreover, the optimization dynamics couple the spurious and signal features, with a stronger spurious component inhibiting signal feature learning. Our analysis reveals precise phase transitions in the learning dynamics. In the first phase, alignment between the signs of the spurious feature and second-layer weight drives rapid growth of the spurious feature. In the second phase, large majority group margin slows learning and the signal feature remains suppressed. When the spurious correlation is maximally strong, we show theoretically that the spurious feature dominates even at the sample complexity threshold where XOR would be learned in isolation (i.e., if the spurious feature was absent). In contrast, when the correlation strength is constant, we provide preliminary empirical evidence that the model can eventually learn the XOR signal, although the spurious feature is not forgotten.
Science Journals
arXiv:2607.04894v1 Announce Type: new Abstract: Memory-based anomaly detection is attractive because it localizes defects from normal images without training a decoder or synthesizing pseudo anomalies. However, most memory methods still use the memory bank as a nearest-neighbor lookup table: a test patch is treated as normal if it has one nearby normal anchor. This hard retrieval view is vulnerable to false-normal matches and does not test whether the patch is consistently supported by a local normal neighborhood. We propose ProCon, a training-free framework that turns memory retrieval into decoder-free reconstruction. ProCon softly projects each test patch onto nearby normal memory vectors and uses the projection residual as anomaly evidence. To stabilize this residual, it constructs seed-perturbed layer-wise memories, aggregates bank residuals by a median, and fuses depth-specific residual maps by layer consensus. ProCon requires no decoder training, backbone fine-tuning, learned fusion weights, or pseudo-anomaly supervision. Across MVTec-AD, VisA, and Real-IAD under the single-category evaluation protocol, ProCon achieves strong image- and pixel-level performance under seven standard metrics, including image AUROC scores of 99.8%, 99.2%, and 93.2%, respectively. Ablations show that the gains come from replacing hard retrieval with soft normal projection and stabilizing the residuals through memory and depth consensus. The code is available at https://github.com/jw-chae/Procon
arXiv:2607.04898v1 Announce Type: new Abstract: We study 3D point cloud reconstruction from multiple partially observed 2D projections. Given two or more projections of an unknown 3D point cloud, together with cross-view point correspondences and visibility information, our goal is to recover a consistent 3D configuration when different views contain different subsets of points. We propose 3D Multi-Perspective Embedding (3DMPE), an optimization-based, training-free method that reconstructs the 3D point cloud and, in the variable-projection setting, jointly estimates the projection maps. 3DMPE extends Multi-Perspective Simultaneous Embedding to accommodate missing points and incomplete pairwise distance information across views. We consider both fixed-projection and variable-projection settings. Unlike learning-based reconstruction methods that infer shape from raw images and often depend on training data, 3DMPE operates on geometric observations with established correspondences and does not require category-specific training. Experiments on ShapeNet and Pix3D evaluate reconstruction quality using Chamfer Distance, Earth Mover Distance, and RMSE-Optimize-Align (ROA), and examine the effects of initialization, the number of views, point visibility, and several noise regimes, including noisy distances and erroneous correspondences. The results demonstrate that 3DMPE can effectively reconstruct point clouds from partial multi-view geometric observations.
arXiv:2607.04907v1 Announce Type: new Abstract: Deploying Large Language Models (LLMs) in high-stakes clinical settings remains limited by structural hallucinations, weak deterministic reasoning over tabular patient data, and omissions in vector retrieval. This paper presents the architecture and validation of Medi-Gemma, a Clinical Decision Support System (CDSS) for wound pathology triage and workflow automation. The platform introduces a decoupled framework that separates clinical perception from data orchestration while preserving traceable reasoning. Medi-Gemma uses a multi-stage pipeline coordinated by a centralized ClinicalOrchestrator. Data requests are handled without generative inference by a DataManager that cleans unstructured Electronic Medical Record (EMR) files through type coercion. Natural language queries are processed by a hierarchical IntentRouter, which routes requests to deterministic analytics paths executed by a PandasQueryEngine or to patient-specific reasoning managed by a ClinicalRAGEngine using a CPU-optimized vector store. A key contribution is the Ground Truth Injection Module, which intercepts patient-specific queries, extracts numeric identification tokens, queries the structured dataframe via Pandas, retrieves the latest validated clinical state, and embeds this snapshot as an overriding context block in the LLM prompt before generation. Safety compliance is enforced by a deterministic ProtocolManager that maps clinical terminology to fixed evidence-based risk pathways, while a SafetyVerifier phrase filter prevents output rule violations. Validation shows that this architecture eliminates semantic context drift, prevents database compilation crashes, and improves factual adherence to backend clinical repositories. These results support Medi-Gemma as a safer pattern for LLM-based clinical decision support where structured data fidelity, retrieval grounding, and deterministic safeguards are essential.
arXiv:2607.04190v1 Announce Type: new Abstract: Global reanalysis products such as ERA5-Land provide spatially complete weather fields but at resolutions too coarse for local applications, particularly in mountainous regions where temperature can vary by several degrees over short distances. This project investigates Convolutional Conditional Neural Processes (ConvCNPs) for statistical downscaling of daily maximum temperature from the ~11km resolution ERA5-Land grid to ~1km resolution over Switzerland, building upon the architecture of Vaughan et al. (2022) and adapting it to the topographically complex Swiss domain with high-resolution elevation features from the swisstopo DHM25. The best model, trained on ten years of data (2014-2023) with five-fold temporal cross-validation, achieves a mean absolute error of 1.31 Celsius and a CRPS-based skill score of 0.524 relative to bilinear interpolation, reducing the expected prediction error by more than half. An ablation study reveals that the elevation MLP is the indispensable component - without it, the model diverges entirely - while explicit seasonal features and Topographic Position Index provide secondary benefits. Under sparse on-grid input the model degrades gracefully, maintaining positive skill down to approximately 10% of the input grid; however, zero-shot deployment on off-grid station observations does not achieve positive skill at any density tested. All configurations exhibit severely overconfident uncertainty estimates, a structural limitation of the Gaussian likelihood training objective. These results demonstrate that ConvCNPs are a viable and effective approach to climate downscaling in complex terrain, and identify uncertainty calibration and native support for non-gridded input as the key challenges for operational deployment.
arXiv:2607.04923v1 Announce Type: new Abstract: The diagnosis of spinal diseases is often assisted by 3D imaging techniques in clinical practice. However, precise 3D spinal assessment is limited by the high costs of 3D imaging hardware and the challenges posed by the physical differences between imaging modalities, which hinder the generalizability of models. To address these issues, we propose UniSpine-GS, an efficient, physics-aware Gaussian framework designed for novel-view projection rendering in multi-view spine imaging via a 3D-aware representation. Instead of performing explicit 3D reconstruction, our approach learns a geometry-aware Gaussian representation that ensures anatomical consistency across different views. We introduce SPWM, a structure-guided loss reweighting strategy to improve boundary fidelity and local details. We evaluate our method on the CTSpine3D dataset and a newly constructed 3D fetal ultrasound dataset, FeSpine3D. Our results demonstrate that UniSpine-GS significantly outperforms existing methods across all metrics, offering a practical and cost-effective solution for unified multi-view medical imaging. Our code is publicly available at https://github.com/orangeisland66/UniSpine-GS.
arXiv:2607.05032v1 Announce Type: new Abstract: Background: Disease severity is a multidimensional construct difficult to capture with rule-based approaches in Electronic Healthcare Records (EHR). Agentic large language model (LLM) systems could synthesise clinical evidence and reason over EHRs, but remain unevaluated for this task. Methods: MOSAIC is a two-phase agentic LLM framework for severity phenotyping, using type 2 diabetes (T2D) as a proof-of-concept. MOSAIC was evaluated on a synthetic cohort (SyntheticMass; open-weight N = 4,886; closed-weight N = 200) against three algorithmic ground truths (DCSI, DiSSCo, Cooper) and against all-cause mortality and incident complications. Open-weight (locally deployable) and proprietary pipelines were also compared. Results: The generated framework spanned domains absent from the comparators, including biomarker-based glycaemic staging, beta-cell function, and social determinants of health. Open-weight MOSAIC matched the proprietary pipeline (closed- vs open-weight weighted kappa = 0.773) and reached moderate agreement with Cooper (kappa = 0.597) and DCSI (kappa = 0.534) and fair agreement with DiSSCo (kappa = 0.320). Agent-based (Type 1) tiers showed significant separation of all-cause mortality (log-rank p < 0.001; crude hazard ratios 1.6-2.4 for non-Baseline tiers), with non-monotonic separation at the upper tiers, and an inverse gradient for incident complications (log-rank p < 0.001) consistent with depletion of susceptibles. Agentic classification also diverged from deterministic execution of the same rubric (MOSAIC Frozen; kappa = 0.428), indicating reasoning beyond fixed rules. Conclusion: MOSAIC shows agentic LLM systems can generate and apply clinically meaningful severity phenotypes from structured EHR data in T2D. Extending it to other diseases with similarly multidimensional severity warrants further research.
arXiv:2607.04924v1 Announce Type: new Abstract: The automotive industry runs on a dense, standardised chain of supplier-quality and certification evidence: production part approval packages, initial sample reports, material certificates, inspection sign-offs, and the type approval dossiers. The chain is operationally central, yet no literature maps it as an information system, and its failures are studied as corporate misconduct rather than system outcomes. This paper systematises the chain and taxonomises its documented failures. First, it systematises the supplier-quality evidence chain across the two dominant regimes, AIAG PPAP and VDA Volume 2 PPA, organised by artefact, producer, verifier, trigger, approval state, and retention. Second, it compiles a compendium of thirteen compliance-documentation failures made public 2012 to 2024, built strictly from the public record and each classified by evidentiary status (adjudicated, company-acknowledged, or alleged), with the Japanese certification cluster of 2016 to 2024 as centrepiece. Third, it builds a failure-mode taxonomy over the compendium, dimensioned by mechanism, lifecycle locus, driver, and detection path, shown exhaustive and discriminating over the case set. Fourth, it derives evidence architecture requirements mechanism by mechanism and sets an open-problems agenda. Across all thirteen cases, not one failure surfaced through the chain's own routine verification; the record indicts the verification layer, not only the evidence authors.
arXiv:2607.04223v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) reduces but does not eliminate hallucination, and existing detectors return a single answer-level score that does not indicate which sentence is unsupported, or why. To close this gap, we introduce Grounding-Aware Sensitivity by Perturbation (GASP), a span-level detector that scores each answer sentence by how strongly its likelihood depends on the retrieved evidence, a quantity we term grounding sensitivity. GASP holds the answer fixed and re-scores it under the full context, under no context, and with each chunk removed, then measures the log-likelihood drops and Jensen-Shannon divergences (JSD). The likelihood of a grounded sentence collapses once its supporting passage is removed, whereas a hallucinated sentence is almost unaffected, a contrast we interpret by casting decoding as a random nonlinear iterated function system (RNIFS). We evaluate GASP on three benchmarks (RAGTruth, TofuEval, RAGBench) with three instruction-tuned scorers from two model families (Qwen2.5-0.5B, Qwen2.5-1.5B, and SmolLM2-1.7B) under a leakage-clean protocol. On RAGTruth it reaches a response-level area under the ROC curve (AUC) of about 0.73 and a span-level AUC of about 0.67, improving significantly over perplexity and by clear margins over length, whole-context natural language inference (NLI), and self-consistency baselines. The only baseline competitive at the span level is a well-configured chunk-level entailment verifier, which requires a separate model, whereas a training-free threshold on the grounding features matches the trained classifier without labeled data and serves as the default detector. Beyond RAGTruth, the signal transfers to TofuEval but not to short-answer question answering in RAGBench, showing GASP is best suited to outputs constructed from the retrieved context rather than answers recoverable from parametric knowledge.
arXiv:2607.04930v1 Announce Type: new Abstract: In the pursuit of robust and generalizable category-level object pose estimation, most existing methods adopt parametric formulations that learn effective representations from data, yet they primarily encode category-level patterns into fixed shape priors or static parameter weights, which limits their scalability to highly diverse instances. In this paper, we rethink category-level pose estimation from a memory-centric perspective and present MemPose, a memory-augmented framework that explicitly incorporates category-level geometric memory into the pose estimation pipeline. We introduce an external memory buffer that stores and dynamically updates structural representations from previously observed instances, enabling the model to leverage accumulated experience to support current perception. Extensive experiments on four challenging benchmarks (REAL275, CAMERA25, Housecat6D and Wild6D) demonstrate the superiority of our proposed method over previous state-of-the-art approaches.
arXiv:2507.02950v3 Announce Type: replace Abstract: Large language models (LLMs) may support counseling training, yet evidence from Japanese-language interactions and automated quality ratings remains limited. We examined 18 fixed Japanese-language counseling transcripts generated through artificial intelligence (AI)-to-AI interactions under three counselor conditions: GPT-minimal (GPT-4-turbo with a minimal role instruction), GPT-SMDP (GPT-4-turbo with the Structured Multi-step Dialogue Prompt [SMDP]), and Claude-SMDP (Claude-3-Opus with SMDP). Fifteen counseling experts rated transcripts on four adapted global scales from the Motivational Interviewing Treatment Integrity coding manual and an overall-quality item; three newer LLMs independently rated the same transcripts in three iterations. In this fixed stimulus set, SMDP-condition dialogues received higher expert ratings for cultivating change talk, partnership, empathy, and overall quality than GPT-minimal dialogues; the two SMDP counselor models did not differ. LLM ratings were reproducible but generally more lenient than expert-reference ratings, particularly for softening sustain talk and overall quality. Simulated-client naturalness was below the scale midpoint. These findings provide an expert-referenced benchmark for Japanese-language AI counseling simulations and show that reproducible LLM ratings should not be treated as calibrated counseling-quality evidence without expert validation. This study does not test clinical effectiveness or human-client outcomes.
arXiv:2607.04951v1 Announce Type: new Abstract: Standard distributed \ac{llm} schedulers rely on static token counts or rolling latency averages, making them susceptible to failures on statutorily constrained text. On \ac{epo} claims governed by Article 84 \ac{epc}, linguistic rigidity makes human and machine authorship statistically indistinguishable. Resolving this ambiguity mid-flight forces dynamic multi-model ensemble expansion, triggering unpredictable KV-cache and weight-allocation spikes that saturate consumer-grade edge GPU VRAM and cause severe \ac{oom} crashes. To prevent hardware collapse, we propose a CPU-side Linguistic Resource Forecasting (LRF) gateway. The gateway extracts a 16-dimensional text-structure vector and applies an XGBoost predictor to forecast trap-band membership. The resulting escalation probability ($\Pesc$) is evaluated against a dynamic, closed-form routing threshold ($\Tauroute(t)$) computed via real-time latency telemetry. Requests are safely routed to either a local Qwen2.5-7B edge worker or a remote contrastive ensemble (Qwen2.5 7B + 32B) on an NVIDIA H100 \emph{before} any edge GPU memory is allocated. In a 6,000-request live trial, the LRF gateway reduced the operational misroute fraction ($R_{\mathrm{mis}}$) to $0.087$--$0.095$, an order of magnitude below the token-count baseline ($0.849$). Peak edge VRAM remained safely bounded at $\SI{4.82}{\gibi\byte}$ (under the $\SI{8}{\gibi\byte}$ ceiling) across a $27\times$ variation in \ac{wan} delay. The predictor achieved a live-trial AUROC of $0.84$, and the dynamic $\Tauroute(t)$ controller yielded an $8.2\%$ relative reduction in misroutes compared to an equivalent static threshold.
arXiv:2607.04962v1 Announce Type: new Abstract: Conspiracy theories commonly attribute important events to the actions of powerful and secretive actors. While computational research has largely focused on document-level analyses of conspiracy theories, less attention has been paid to identifying the actors that drive such narratives. We develop annotation guidelines for conspiratorial actors, present a span-annotated corpus of German Telegram posts, and investigate their automatic extraction using transformer-based models. We further apply the resulting model to the \textit{Schwurbelarchiv}, a large-scale archive of German conspiracy-related Telegram channels. Our results demonstrate that conspiratorial actors can be annotated with meaningful agreement and extracted with reasonable accuracy despite the linguistic complexity of conspiracy discourse, enabling large-scale analyses of actor representations in conspiracy narratives.
arXiv:2606.31753v2 Announce Type: replace-cross Abstract: Recently, Doosti et al. introduced the problem of distributed quantum state verification, where $m$ distributed nodes are given a copy of an unknown state $\rho$, and can send limited one way communication to a central node, who has a complete description of a known state $\sigma$. They ask how many distributed nodes $m$ are required, before the central node can succeed at distinguishing whether $\rho=\sigma$ or $\|\rho-\sigma\|_1\geq\varepsilon$ with high probability. In the setting where only quantum communication is allowed, Doosti et al. exhibit conditional lower bounds in both the public and private-coin settings, and a matching upper bound in the public-coin setting. We extend these results, and show unconditional lower bounds for when both classical and quantum communication are permitted. We show the public-coin lower bound is tight by giving an algorithm with a matching upper bound. We also show an almost tight upper bound in the private-coin setting when only quantum communication is permitted.
arXiv:2607.05015v1 Announce Type: new Abstract: Thermodynamic integration (TI) is a widely used approach for computing free energies and phase diagrams. However, TI calculations driven by machine learning interatomic potentials (MLIPs) remain technically challenging because they require careful design of reversible integration paths and many closely related molecular dynamics (MD) tasks for each phase and state point. To address these challenges, we present dpti, an open-source Python package that automates TI workflows for phase diagram calculations with MLIPs. dpti connects reference systems with analytically known free energies to MLIP-described atomic and molecular solids and liquids through reversible integration paths. Given JSON input files, dpti generates and runs the required MD tasks, computes free energy contributions, estimates errors, and propagates coexistence points into phase boundaries. We demonstrate the usage of dpti with two examples driven by Deep Potential models: a silica phase diagram involving beta-quartz, coesite, and melt, and the ice Ih-liquid water phase boundary. dpti provides a useful tool for automated phase diagram calculations of materials modeled by MLIPs.
arXiv:2607.04986v1 Announce Type: new Abstract: This article tackles an important phenomenon in the syntax of Yemeni Ibbi Arabic (YIA), viz., wh-agreement, a phenomenon common to several languages including Greek, Indonesian, Lubukusu, Irish, etc. In YIA, wh-agreement manifests itself via agreement inflections on the Wh-Op, C, T/V, v. To account for this phenomenon, we propose an Agree across phases (AAP) approach anchored in the mechanism of Feature Inheritance (FI) in which Agree as MATCHING (AM) is a bit separated from feature valuation (FV). AM concerns Cs/vs, but FV Ts/Vs. Analyzing the agreement patterns observed between Wh-Op(erators), functional heads (precisely C, (T), v), and verbal complexes, we argue that the suffixes -eh, -uh, -nen, -um, having undergone grammaticalization process from Stannard Arabic (SA) third person pronouns, function as morphological marking of wh-agreement. Findings indicate that YIA data offer a unique empirical contribution to generative syntax, specifically concerning wh-agreement in this dialect operating via MATCHING mechanism. Our proposal straightforwardly accounts for wh-agreement cross-linguistically. This study provides further evidence that incorporating under-investigated typology provides further support for the universality of Universal Grammar (UG) by revealing how specific I-language operations reflect deeper, invariant principles of human language architecture. It concludes that the wh-agreement mechanism in YIA is more morphosyntactically robust than in languages such as Greek, Indonesian, Palauan, and Irish, providing compelling evidence for AAP as a UG approach to long-distance dependencies.
arXiv:2607.04989v1 Announce Type: new Abstract: We present a method for constructing intrinsic triangulations of closed discrete surfaces, in which edges correspond to shortest geodesic paths and faces decompose into geometric primitives inherited from the underlying mesh. Starting from a watertight input triangulation, the method progressively builds an intrinsic mesh through local optimization operations -- edge swaps, edge splits, edge collapses, and triangle splits -- performed directly on the surface without modifying the original geometry. Element size is controlled via a characteristic length field, and quality is enforced through angle-based criteria derived from intrinsic distances. Geodesic distances are computed exactly using a continuous Dijkstra approach, accelerated by an A* search strategy that reduces computation to roughly $3\%$ of the cost of standard propagation. The framework supports both refinement and coarsening, overcoming a key limitation of prior intrinsic methods based on developable triangles. As a by-product, the intrinsic triangulation provides a natural foundation for direct high-order mesh generation, bypassing the classical pipeline of first constructing a linear mesh and subsequently curving it. The method is validated on the Thingi10K dataset across nearly 5,000 geometrically complex models.
arXiv:2507.10058v2 Announce Type: replace Abstract: Optical tweezers have become a standard tool for manipulating microscale and nanoscale particles and probing their local environments. However, complex particle dynamics under optical forces typically require structured light fields, multi-beam traps, or engineered environments. Here we achieve complex particle dynamics in a single Gaussian-beam optical tweezer. The effect originates from higher-order Mie resonances supported by wavelength-scale particles. In our optical tweezer, small particles in the regime of Rayleigh scattering or the lowest-order dipole-type Mie modes remain confined at the beam center. By contrast, particles within the range of sizes corresponding to quadrupole-type Mie modes exhibit more complex behavior. In a linearly polarized Gaussian beam, these particles are trapped in a potential with two off-axis equilibria. We observe thermally driven hopping between these equilibria, with the hopping frequency controlled by the laser power. In a circularly polarized Gaussian beam, the particles are confined to a stable orbit and exhibit circular motion driven by the spin (circular-polarization) degree of freedom of the beam, with angular velocity dependent on the laser power. These results reveal higher-order Mie resonances as an intrinsic mechanism behind complex optical forces. This establishes Mie-resonant nanophotonics as a flexible platform for inducing and controlling complex motion in optical tweezers for nanoparticle manipulation as well as sensing of local environments.
arXiv:2509.07218v5 Announce Type: replace Abstract: The rapid growth of artificial intelligence (AI) is driving an unprecedented increase in the electricity demand of AI data centers, raising emerging challenges for electric power grids. Understanding the characteristics of AI data center loads and their interactions with the grid is therefore critical for ensuring both reliable power system operation and sustainable AI development. This paper provides a comprehensive review and vision of this evolving landscape. Specifically, this paper (i) presents an overview of AI data center infrastructure and its key components, (ii) examines the key characteristics and patterns of electricity demand across the stages of model preparation, training, fine-tuning, and inference, (iii) analyzes the critical challenges that AI data center loads pose to power systems across three interrelated timescales, including long-term planning and interconnection, short-term operation and electricity markets, and real-time dynamics and stability, and (iv) discusses potential solutions from the perspectives of the grid, AI data centers, and AI end-users to address these challenges. By synthesizing current knowledge and outlining future directions, this review aims to guide research and development in support of the joint advancement of AI data centers and power systems toward reliable, efficient, and sustainable operation.
arXiv:2607.04975v1 Announce Type: cross Abstract: What distinguishes front from back? In physics, such directionality emerges only when an underlying symmetry is broken. Antiferromagnets that inherently break both space-inversion and time-reversal symmetries provide a striking example, exhibiting nonreciprocal optical responses that depend on the direction of light propagation. Beyond distinguishing antiferromagnetic domains, we show that this nonreciprocity can deterministically create them. Using mid-infrared light, we demonstrate deterministic switching of antiferromagnetic domains in the magnetoelectric antiferromagnet LiFePO4, where illumination from opposite sides selectively stabilizes opposite domain states. Remarkably, the switching persists over a broad wavelength range rather than being confined to a narrow transition-specific spectral region, overcoming the spectral and material constraints of resonance-based optical switching schemes. The broadband switching originates from the material's intrinsic nonreciprocity through optically generated heat currents. Our results establish nonreciprocity as a general principle for deterministically controlling symmetry-broken phases with light.
A Systematic Survey on Large Language Models for Evolutionary Optimization: From Modeling to Solving
arXiv:2509.08269v5 Announce Type: replace Abstract: Large language models (LLMs) are increasingly integrated with evolutionary computation to support optimization tasks. This survey primarily focuses on evolutionary optimization, i.e., optimization based on evolutionary computation. For brevity, we use the term optimization throughout to denote this scope. However, existing surveys typically examine isolated roles of LLMs and do not provide a unified view that connects optimization modeling with optimization solving. To address this gap, we systematically review recent developments through a workflow-oriented framework. First, we organize the literature into two primary stages: LLMs for optimization modeling and LLMs for optimization solving (in this survey, the terms optimization modeling and optimization solving are used as concise forms of optimization problem modeling and optimization problem solving, respectively). Second, we divide the solving stage into three paradigms according to the role of the LLM: stand-alone optimizers, low-level components embedded in optimization algorithms, and high-level managers for algorithm selection and generation. Third, we analyze representative methods, identify their technical limitations, and clarify their relationships with traditional optimization approaches. We further substantiate this taxonomy through benchmark systematization, baseline comparisons, and practitioner-oriented guidance, and we review interdisciplinary applications across the natural sciences, engineering, and machine learning. Based on the resulting analysis, we identify research directions toward dynamic, self-evolving, and agentic optimization ecosystems. An up-to-date collection of related literature is maintained at https://github.com/ishmael233/LLM4OPT.
arXiv:2509.22267v5 Announce Type: replace Abstract: Reliable detection of bearing faults is essential for maintaining the safety and operational efficiency of rotating machinery. While recent advances in machine learning (ML), particularly deep learning, have shown strong performance in controlled settings, many studies fail to generalize to real-world applications due to methodological flaws, most notably data leakage. This paper investigates the issue of data leakage in vibration-based bearing fault diagnosis and its impact on model evaluation. We demonstrate that common dataset partitioning strategies, such as segment-wise and condition-wise splits, introduce spurious correlations that inflate performance metrics. To address this, we propose a rigorous, leakage-free evaluation methodology centered on bearing-wise data partitioning, ensuring no overlap between the physical components used for training and testing. Additionally, we reformulate the classification task as a multi-label problem, enabling the detection of co-occurring fault types and the use of prevalence-independent metrics based on the ROC curve. Beyond preventing leakage, we also examine the effect of dataset diversity on generalization, showing that the number of unique training bearings is a decisive factor for achieving robust performance. We evaluate our methodology on four widely adopted datasets: Case Western Reserve University (CWRU), Paderborn University (PU), University of Ottawa (UORED-VAFCLS) and Hanoi University of Science and Technology (HUST bearing). This study highlights the importance of leakage-aware evaluation protocols and provides practical guidelines for dataset partitioning, model selection, and validation, fostering the development of more trustworthy ML systems for industrial fault diagnosis applications.
arXiv:2601.20841v4 Announce Type: replace Abstract: The Reynolds equation is derived from the incompressible Navier Stokes equations under the lubrication assumptions of a long and thin domain geometry and a small scaled Reynolds number. The Reynolds equation is an elliptic differential equation and a dramatic simplification from the governing equations. When the fluid domain is piecewise linear, the Reynolds equation has an exact solution that we formulate by coupling the exact solutions of each piecewise component. We consider a formulation specifically for piecewise constant heights, and a more general formulation for piecewise linear heights; in both cases the linear system is inverted using the Schur complement. These methods can also be applied in the case of non-linear heights by approximating the height as piecewise constant or piecewise linear, in which case the methods achieve second-order accuracy. We assess the time complexity of the two methods, and determine that the method for piecewise linear heights is linear time for the number of piecewise components. As an application of these methods, we explore the limits of validity for lubrication theory by comparing the solutions of the Reynolds and the Stokes equations for a variety of linear and non-linear textured slider geometries.
arXiv:2607.04478v1 Announce Type: new Abstract: Automated chest X-ray classification remains challenging due to severe class imbalance, co-occurring pathologies, and the loss of localized features in conventional architectures. To address these, we propose an explainable hierarchical multi-view ensemble framework for the robust classification of 14 thoracic pathologies. The framework employs view-specific training by independently modeling frontal and lateral radiographs using an ensemble of five complementary convolutional neural networks. Replacing global average pooling, a multi-scale feature fusion strategy augmented with Convolutional Block Attention Modules (CBAM) preserves fine-grained intermediate representations while emphasizing high-level pathology-specific semantic features. To mitigate positive-negative imbalance and varying inter-class difficulty, models are optimized using a novel hybrid objective combining Asymmetric Loss with Adaptive Focal Loss. Beyond simple probability averaging, the framework incorporates a hierarchical meta-learning strategy where test-time augmentation (TTA) predictions and cross-model uncertainty measures are integrated into Level-1 gradient-boosting meta-learners (XGBoost, LightGBM, and CatBoost), followed by Level-2 stacking with optimized alpha blending. Evaluated on a large-scale CheXpert-style dataset, the framework achieves state-of-the-art macro-average AUROC scores of 0.9319 for frontal and 0.9154 for lateral radiographs. Furthermore, comprehensive explainability analysis using seven post-hoc attribution techniques demonstrates strong anatomical consistency and clinically meaningful decision localization. By integrating architectural diversity, multi-scale attention, hierarchical meta-learning, and rigorous explainability, the proposed framework provides a transparent, highly accurate, and clinically practical computer-aided diagnosis system for thoracic disease classification.
arXiv:2601.22901v2 Announce Type: replace Abstract: In this paper, we study how sensing and communication should be jointly coordinated in integrated sensing and communication (ISAC) systems to maintain timely situational awareness under reliability and resource constraints. We consider an ISAC-enabled base station that supports a remote source by dynamically choosing between sensing new state information and communicating previously acquired information, with the two operations semantically intertwined rather than serving separate targets and users. Both sensing and communication are unreliable and costly. The objective is to optimise a long-term cost that captures information freshness at the source, measured by the age of information (AoI), together with sensing and communication overheads. The resulting sequential decision problem is formulated as an infinite-horizon Markov decision process (MDP) with two-dimensional AoI states that capture information freshness at the source and at the base station. We prove that the optimal stationary policy admits a monotone threshold structure characterised by a nondecreasing switching curve in the AoI state space, and show that, as the base-station information becomes staler, the system increasingly favours sensing over communication. Our numerical analysis corroborates the theoretical findings.