arXiv:2607.14944v1 Announce Type: new
Abstract: Recent advances in data-driven modelling have highlighted the potential of hybrid approaches which combine Tensor Basis Neural Networks (TBNN) with Universal Differential Equations (UDE) to discover frame-invariant, non-linear viscoelastic constitutive models. These hybrid models enable the creation of digital twins for complex viscoelastic fluids, offering direct transferability to computational fluid dynamics simulations. In this work, we introduce a reduced dimensional tensor basis formulation that enhances both the physical consistency of the learned representations with respect to the training data and the numerical stability of subsequent simulations. The UDE architecture is embedded into an open-source finite volume solver in which the constitutive response is generated dynamically at runtime based on local fluid flow conditions. Training on synthetic datasets generated using a range of well established viscoelastic models in oscillatory shear flows alone, the performance of the resulting UDEs is evaluated under extrapolation to unseen conditions and flow-types. These include deploying the UDEs in viscometric extensional flows as well as 2D and 3D benchmark flows, such as the 4:1 sudden contraction and cross-slot, providing a quantitative analysis of their capabilities, limitations and failure modes. The proposed reduced-basis framework enables data-efficient discovery of frame-invariant constitutive models that generalise beyond their training regime, capturing key flow features such as the onset and growth of flow-induced elastic instabilities in strong extensional flows even though trained solely on shear data. Quantitative accuracy decreases as extrapolation increases, but incorporating first normal stress difference information further improves quantitative accuracy and extends predictive fidelity to higher Deborah numbers.
Science Journals
arXiv:2607.14968v1 Announce Type: new
Abstract: Surgical video understanding is fundamental to navigation systems. Endoscopic perception is often hindered by a limited field-of-view and frequent instrument occlusions, making spatio-temporal context essential for robust inference. These challenges have motivated video models that aggregate information across frames. However, existing video models typically store past observations implicitly in learned feature representations, often requiring task-specific video training, substantial annotated data, and increased computational cost. We propose Stitch-Inferencer, a real-time, model-agnostic inference framework that replaces implicit feature memory with an explicit image-space panoramic canvas. By stitching valid observations across frames, Stitch-Inferencer preserves previously observed pixels in an online, instrument-free view, expanding the effective field-of-view and providing direct access to regions that are temporarily occluded or absent from the current frame. Downstream segmentation or tracking models are applied to a compact region of interest on the panorama, and their predictions are reprojected to the current frame, enabling existing models to exploit long-range context without retraining. Experiments on anatomy segmentation and point/box tracking demonstrate consistent improvements across diverse baselines while preserving real-time throughput. The stitching module alone runs at over 60 FPS, providing a practical inference-time solution to enhance endoscopic perception in computationally constrained intraoperative environments. Source code will be made publicly available.
arXiv:2607.14970v1 Announce Type: new
Abstract: Automated optimisation is increasingly adopted in industrial processes, yet a trust gap persists between engineers who design these algorithms and operators who must act on their recommendations. Explainable AI methods like SHAP (SHapley Additive exPlanations) have transformed interpretability for machine learning predictions; optimisation outputs could benefit from similar techniques. We present an approach that integrates Implicit Function Theorem (IFT) based sensitivity analysis with SHAP attribution and narrative generation via Large Language Models (LLM), producing explanations tailored for operators. Our approach leverages IFT to compute exact parameter sensitivities $\partial p^*/\partial x$ from the optimality conditions, enabling efficient GradientSHAP computation. For an industrial High Pressure Grinding Roll (HPGR) control optimisation problem with 22 features, we achieve equivalent SHAP attributions (correlation $>$0.99 with KernelSHAP) with over 40$\times$ speedup, enabling real-time natural language explanations. We validate on industrial scenarios and present feedback from domain experts on generated explanations.
arXiv:2607.15271v1 Announce Type: new
Abstract: Online novel view synthesis from multi-view streaming videos faces a fundamental trade-off: maintaining a persistent, long-horizon memory to reconstruct temporarily occluded regions while operating under strict real-time constraints. While Test-Time Training (TTT) offers a powerful memory mechanism, standard models mandate gradient-based memory updates at every frame to adapt to the changing motion in dynamic scenes. The computational cost of heavy memory updates precludes real-time application and can lead to instability over long contexts. Given that memory updates are more demanding than memory application and video content is largely redundant, we propose to decouple the frequencies of these two processes. Our approach performs periodic memory updates while applying the memory on a per-frame basis, using cross-view attention to manage deformations between the prior memory state and the current frame. To lock in the historical context, we introduce two critical mechanisms: an auxiliary Memory Loss that forces persistent internalization of the scene, and a Memory Caching strategy that regularizes active weights against catastrophic drift. Our method demonstrates real-time, state-of-the-art performance on scenes with dynamic human motion as well as minute-scale online memorization.
arXiv:2607.15272v1 Announce Type: new
Abstract: Editing the figures in a research paper is a routine and time-consuming part of everyday research practice: authors relabel components, rearrange panels, and restyle visuals as they revise their manuscripts. Automating this editing workflow under a natural-language instruction, however, is challenging, because a scientific figure is a dense infographic in which heterogeneous visual elements such as schematics, plots, photos, captions, and arrows are composed under a tight visual grammar to advance a specific argument. To address this, we present SciDiagramEdit, a benchmark and skill-evolution framework that learns from natural paper revisions and operates on the figure's editable vector source, where users can inspect and co-edit individual primitives alongside the agent. Our benchmark mines before/after figure pairs from arXiv version histories, each grounded in the authors' own revision intent. To accommodate the diversity of editing instructions, we adopt agentic learning via skill evolution: an agentic proposer continually refines the agent's skill specification from execution traces over multiple epochs. The resulting skill progressively lifts edit accuracy on a held-out validation set, providing evidence that natural paper revisions are an effective training signal for instruction-driven figure editing.
arXiv:2607.14761v1 Announce Type: new
Abstract: Label-free all-optical nanosizing approaches based on interferometric or darkfield-imaging infer size, composition, or shape from single-particle scattering signals, but these signals are inseparably coupled to the spatially non-uniform illumination profile of the imaging system. Existing normalisation strategies require directly measuring this illumination field, an approach that fails for background-free geometries, such as lightsheet-type illumination, where the field cannot be detected. Here we introduce a self-normalisation method that reconstructs the illumination profile directly from the scattering signals of many freely diffusing nanoparticles, requiring no additional hardware, calibration samples, or direct field measurement. Critically, our approach eliminates the particle-heterogeneity bias that otherwise corrupts such reconstructions, by normalising single-particle trajectories against each other in regions of spatial overlap, where distinct particles necessarily sample identical illumination and detection conditions. We validate this method for gold nanoparticles of various size in two- and three-dimensional geometries, including a 90{\deg} side-illumination configuration in which the illumination field is entirely undetectable by conventional means, and show that reconstructed profiles closely match ground-truth measurements, thus drastically reducing signal variability. Relying solely on the scattering signal already acquired for sizing, our approach is immediately compatible with existing interferometric and darkfield nanoscopy platforms and broadly extendable to other scattering or fluorescent modalities, including light-sheet microscopy.
arXiv:2607.15059v1 Announce Type: new
Abstract: The Matilda effect describes the systematic under-recognition of women's scientific contributions. We investigated its historical evolution by analyzing over 220 million publications (1900-2025) from OpenAlex, inferring the gender of roughly 60 million authors. To quantify this under-recognition, we calculated the ratio between the overall share of female authors and female corresponding authors. Because corresponding authorship denotes intellectual leadership and primary academic credit, systematic exclusion from this role directly measures the Matilda effect. Our analysis reveals that overall female participation rose from approximately 10% in the 1910s to over 40% recently. For most of the 20th century, women were disproportionately excluded from corresponding authorship, confirming a historical Matilda effect. However, this trend inverted over the past two decades, indicating a global decline. Despite this overall progress, distinct disparities persist: the physical sciences and Asian countries lag behind, whereas the health sciences and Latin America approach gender parity.
arXiv:2607.15278v1 Announce Type: new
Abstract: Video models are evolving into vision foundation models, yet they still lack human-like multi-step reasoning. Streaming autoregressive diffusion models are efficient but limited in reasoning, while bidirectional diffusion enables global revision with high inference costs due to dense frame-level denoising. Both paradigms struggle to achieve logical consistency and low-latency streaming for complex reasoning tasks. We propose HDR (Hierarchical Denoising for Visual Reasoning), a unified framework that integrates hierarchical latents into causal video generation for multi-step reasoning. HDR organizes video latents into a tree-structured hierarchy, enabling coarse-to-fine reasoning before streaming output. Coarse denoising layers preserve uncertain hypotheses for global planning, while finer layers progressively refine them into concrete visual states. A sparse hierarchical attention pattern (SHAP) further reduces temporal attention costs. We introduce a level-stratified multi-step video reasoning benchmark with out-of-distribution cases, covering six tasks: maze navigation, Tower of Hanoi, one-line drawing, sliding puzzle, Sokoban, and water pouring. Compared with streaming autoregressive diffusion baselines, HDR improves success from 34.22 to 60.29 (76.2% relative gain) and increases average progress from 76.00 to 89.56, demonstrating more consistent reasoning trajectories. HDR maintains low-latency streaming at 0.70 seconds per latent, achieving 54.2 times faster inference than bidirectional diffusion. It also retains 82.9% of full-data performance with only 2% training data, compared with 52.0% for bidirectional diffusion. Real-world robot experiments further demonstrate HDR's potential for physical interaction and world modeling. Project demo: https://hierarchical-diffusion-reasoning.github.io/.
arXiv:2607.15198v1 Announce Type: cross
Abstract: We introduce the REAL-TSE Challenge, an IEEE SLT 2026 satellite challenge on target speaker extraction~(TSE) from real conversational recordings. Given a multi-speaker mixture and one or more enrollment utterances from a target speaker, participating systems must recover only the target speech. Unlike simulated read-speech benchmarks, REAL-TSE evaluates Mandarin and English recordings that contain natural overlap, reverberation, noise, channel mismatch, and conversational dynamics. The challenge defines two complementary tracks: an Online track for low-latency streaming extraction and an Offline track for full-context processing. Systems are evaluated with Token Error Rate (TER), Speaker Similarity (SpkSim), DNSMOS, and target-speaker activity F1. This overview paper describes the task definition, datasets, baselines, evaluation protocol, submitted systems, condition-wise findings, and lessons for future real-world TSE benchmarks.
arXiv:2607.15204v1 Announce Type: cross
Abstract: Volatile outgassing from planetary interiors controls the composition of rocky exoplanets' secondary atmospheres. However, observations indicate that disequilibrium processes, such as photochemistry and vertical transport, can strongly alter the chemical structure of Hot Jupiters. Which process dominates under different types of rocky planets, and how outgassing and photochemistry jointly determine the atmospheric composition, remain open questions. Sulfur species are promising tracers of interior-atmosphere coupling because their atmospheric abundances are sensitive to both mantle redox state and stellar irradiation. The PROTEUS planetary interior-atmosphere evolution modelling framework is coupled to two chemical models, FastChem and VULCAN, for post-processed chemistry calculations. We run a grid of planetary evolution simulations spanning diverse mantle redox states, instellation fluxes, and Solar versus M-star host-star spectra. For each case, we compare atmospheric compositions under thermochemical equilibrium, only vertical transport, and vertical transport plus photochemistry. The bulk atmospheric composition remains controlled by the redox state of the mantle and outgassing history, even when disequilibrium chemistry is included. Reduced mantles produce atmospheres rich in H2, and oxidised mantles are dominated by CO2. Photochemistry affects the upper atmosphere, strongly depleting neutral volatiles and enhancing radicals, especially for highly irradiated cases. SO2 is strongly enhanced at intermediate-to-oxidised redox states. Synthetic emission spectra show that photochemical SO2 can generate absorption features at 4 um and at 7.3 / 8.7 um, reaching ~60 ppm and ~100 ppm, before sequentially returning to the outgassed signatures of ~30 ppm and ~50 ppm for the oxidised mantle redox state. These signatures are detectable with JWST, motivating targeted observational campaigns.
arXiv:2607.14527v1 Announce Type: cross
Abstract: Stein variational gradient descent (SVGD) transports interacting particles toward a target distribution through deterministic kernelized dynamics. Singular Riesz kernels are attractive because they can provide quantitative population-level convergence, but at the finite-particle level the corresponding Stein energy has infinite self-interaction. We study periodic Riesz SVGD with self-interaction removed and prove a many-particle, long-time sampling theorem. Throughout the range in which the singular Stein energy is locally integrable, under a uniform bound on the initial relative entropy per particle, the time-averaged empirical-measure law converges weakly to the point mass \(\delta_\pi\) at the target as the particle number and any diverging averaging horizon tend to infinity. We also show that the empirical-measure laws induced by invariant particle laws of finite relative entropy converge weakly to \(\delta_\pi\), without a uniform entropy bound. Below the logarithmic singularity threshold, we obtain an explicit algebraic finite-particle error bound. These results extend the joint-entropy approach for smooth-kernel SVGD to singular interactions.
arXiv:2607.15066v1 Announce Type: new
Abstract: Dissociative chemisorption (DC) of H$_2$ on the Cu(111) surface is a prototypical problem for understanding elements of heterogeneous catalysis [Science 326, 832 (2009)]. The challenge lies in modeling the reaction dynamics that in turn reflects a classical potential for atomic deformations, friction, and inelastic scattering. Here, I test the use of a set of range-separated hybrid (RSH) van der Waals density functionals (vdW-DFs) [JPCM 37, 211501 (2025)] on their ability to describe the classical barrier for dynamics in this H$_2$+Cu(111) DC problem. I furthermore document use of a variant for fast accurate predictions of the molecular quasi-particles (QPs), finding excellent performance across a set of small molecules that are often studied in catalysis. Finally, I suggest and implement a way to use that QP focus to identify what I consider a best-possible non-empirical (yet adsorbate specific) RSH vdW-DF version, denoted AHBR($\gamma^*$) for H$_2$ DC modeling, navigating what are partly conflicting requirements on the molecule and metal sides. I find that the AHBR($\gamma^*$) can determine the classical H$_2$+Cu(111) DC barrier height close to chemical accuracy. I suggest that DC modeling can test broader relevance of the physics underpinning these RSH vdW-DFs.
arXiv:2607.15068v1 Announce Type: new
Abstract: High-level synthesis (HLS) raises the abstraction level of FPGA accelerator design from hardware description languages to C/C++, but high-quality results still depend on schedule decisions such as pipelining, unrolling, tiling, reordering, and buffering. These decisions create a combinatorial design space, while many numerical kernels exhibit recurring computation patterns that suggest different optimization strategies. This paper presents PATTERNDSE, a lightweight pattern-guided design space exploration (DSE) framework for FPGA kernels written in Allo, a scheduling-oriented HLS programming system. PATTERNDSE maps recurring computation patterns, including elementwise maps, reductions, matrix-vector operations, matrix-matrix operations, and stencil-like updates, to compact schedule spaces. It then applies candidate schedules, validates functional correctness through LLVM execution, checks HLS C code generation, and uses a simple pattern-aware estimator to rank candidates before Vitis HLS synthesis. We evaluate PATTERNDSE on six representative kernels: vecadd, axpy, dot, matvec, gemm, and jacobi2d. Compared with an exhaustive-lite baseline, pattern-guided DSE reduces the number of HLS-evaluated candidates from 140 to 29, achieving a 4.83x overall search reduction and up to 12.0x reduction for individual kernels. Across all evaluated kernels, PATTERNDSE recovers the same best valid Vitis HLS latency as the exhaustive-lite baseline, demonstrating that computation-pattern information can prune unproductive schedule combinations while preserving high-quality HLS outcomes.
arXiv:2607.15225v1 Announce Type: new
Abstract: We present campaign diagrams, a visualization technique for phase-level analysis of resource utilization and bottlenecks in modern workloads. Existing tools have a trade-off: rooflines aggregate a workload into a single point and lose all notion of time, while profilers and traces expose fine-grained events but obscure what bounds performance. Instead, a campaign diagram depicts compute throughput and memory bandwidth utilization, compute and memory traffic volume, and latency in a single figure. Since they can be generated from analytical models, simulations, or profiling data, campaign diagrams capture both ideal bounds and a kernel's achieved performance. We demonstrate them on two case studies: a low-rank GEMM, where they reveal the counterintuitive result that reducing operational intensity can improve end-to-end performance, and Mamba, where they expose fusion and pipelining opportunities across phases. In both cases, our visualization technique reveals optimization opportunities that are difficult to identify with rooflines or profilers alone.
arXiv:2607.14145v1 Announce Type: new
Abstract: Tool-augmented large language model agents excel at long-horizon tasks, yet they are typically post-trained on fixed toolsets. When tasks demand new tools, these agents struggle to incorporate them effectively, and retraining from scratch is often impractical. We identify the core obstacle in such toolset expansion problem as behavioral inertia: the tendency of agents to fall back on familiar tools and established reasoning patterns despite having access to new ones. We demonstrate that injecting counterfactual anchor contexts at critical decision points can break this inertia, recovering failed trajectories by eliciting suppressed agent capabilities. To scale this insight, we propose ToolAnchor, a framework that uses teacher models to hypothesize these counterfactual contexts, verifies them via student rollouts, and internalizes the successful interventions through agentic post-training. Extensive evaluations across general AI assistant (GAIA), textual search (BrowseComp), and visual search (VDR-Bench) tasks demonstrate that ToolAnchor consistently exhibits competitive performance under expanded toolsets. Our work bridges the gap between static post-training and dynamic adaptation, charting a new path for scalable agentic reinforcement learning.
arXiv:2607.14927v1 Announce Type: new
Abstract: While recent flow-matching 3D generative models (e.g., VecSet) adopt structured representations, their tokens share global context, causing conventional training-free editing to suffer from semantic artifacts such as collapsed preserved regions or incomplete transformations. To address this, we propose TanGO, a training-free framework that enables adaptive per-token steering in the tangent space of generative dynamics. To realize this selective control, we formulate a one-step optimal control rule and determine the strength of each token's control signal using a von Mises-Fisher inspired directional discrepancy derived from the source and target velocity fields. Experiments show that TanGO substantially reduces structural artifacts and achieves state-of-the-art performance, outperforming existing 3D editing baselines. The code is publicly available at https://github.com/siw00-lim/TanGO.
arXiv:2511.14441v2 Announce Type: replace-cross
Abstract: To distinguish Markov equivalent graphs in causal discovery, it is necessary to restrict the structural causal model. Crucially, we need to be able to distinguish cause $X$ from effect $Y$ in bivariate models, that is, distinguish the two graphs $X \to Y$ and $Y \to X$. Location-scale noise models (LSNMs), in which the effect $Y$ is modeled based on the cause $X$ as $Y = f(X) + g(X)N$, form a flexible class of models that is general and identifiable in most cases. Estimating these models for arbitrary noise terms $N$, however, is challenging. Therefore, practical estimators are typically restricted to symmetric distributions, such as the normal distribution. As we showcase in this paper, when $N$ is a skewed random variable, which is likely in real-world domains, the reliability of these approaches decreases. To approach this limitation, we propose SkewD, a likelihood-based algorithm for bivariate causal discovery under LSNMs with skewed noise distributions. SkewD extends the usual normal-distribution framework to the skew-normal setting, enabling reliable inference under symmetric and skewed noise. For parameter estimation, we employ a combination of a heuristic search and an expectation conditional maximization algorithm. We evaluate SkewD on novel synthetically generated datasets with skewed noise as well as established benchmark datasets. Throughout our experiments, SkewD exhibits a strong performance and, in comparison to prior work, remains robust under high skewness.
arXiv:2511.06609v4 Announce Type: replace
Abstract: The accurate forecasting of complex, high-dimensional dynamical systems from observational data is a fundamental task across numerous scientific and engineering disciplines. A significant challenge arises from noise-corrupted measurements, which severely degrade the performance of data-driven models. In chaotic dynamical systems, where small initial errors amplify exponentially, it is particularly difficult to develop a model from noisy data that achieves short-term accuracy while preserving long-term invariant properties. To overcome this, we consider the weak formulation as a complementary approach to the classical $L2$-loss function for training models of dynamical systems. We empirically verify that the weak formulation, with a proper choice of test function and integration domain, effectively filters noisy data. This insight explains why a weak form loss function is analogous to fitting a model to filtered data and provides a practical way to parameterize the weak form. Subsequently, we demonstrate how this approach overcomes the instability and inaccuracy of standard Neural ODE (NODE) in modeling chaotic systems. Through numerical examples, we show that our proposed training strategy, the Weak Penalty NODE, is computationally efficient, solver-agnostic, and yields accurate and robust forecasts across benchmark chaotic systems and a real-world climate dataset.
arXiv:2607.14393v1 Announce Type: new
Abstract: Human-in-the-loop Reinforcement Learning has become a popular approach to training, finetuning, and aligning robot behavior with user preferences. Our paper explores the feasibility of using brain signals via functional near-infrared spectroscopy (fNIRS) to modulate robot learning in simulation. We compare agents trained on passive (observational) versus active (demonstrative) interaction tasks, and test multiple methods for enhancing the RL algorithm with the neural signal, focusing on parameter augmentation rather than replacement. We further examine how model granularity and noise affect agent learning. Our results show that this framework is effective: the neural signal improves learning when augmenting trajectory priorities and state-action q-values. Additionally, the framework learns successfully from offline data, offering a practical alternative for settings where real-time BCI setups are impractical or only limited data is available.
arXiv:2607.15242v1 Announce Type: new
Abstract: A common bottleneck in two-stage recommendation is embedding staleness: when a user rates a new item, their embedding remains fixed until the next retrain cycle. We propose mutable sketches, which store each user's preferences in a KP-tree (a sparse segment tree with sum aggregation), fit a low-rank projection once, and recompute embeddings on-the-fly as ratings arrive. We prove that each new observation monotonically tightens the prediction error envelope (Theorem 1), a guarantee that FunkSVD and eALS lack. On KuaiRec, the mutable sketch achieves 0.810 RMSE at 1.8% data read vs. ALS 0.822 at 100%, with 8x faster per-batch updates. A new user receives personalized recommendations in <1 ms after their first rating, with no model retraining required. A comparison of sampling strategies across density regimes shows that the KP-tree's norm-proportional sampling provides 40-130% better item coverage on sparse data (<1% density), while uniform sampling suffices on dense matrices.
arXiv:2607.14268v1 Announce Type: new
Abstract: This paper gives the first convergence analysis and proof of acceleration for nonlinear GMRES (NGMRES) applied to contractive and noncontractive fixed point iterations (FPIs) for solving general nonlinear systems. Our main results are that in both the contractive and noncontractive cases, the ratio gain of the optimization problem is the mechanism responsible for accelerating (or enabling) convergence. Our analysis also reveals a second important quantity related to the optimization problem, which directly predicts the linear convergence rate at each iteration and proves it is at most 1; hence only higher order terms are responsible for NGMRES non-convergence. Numerical results for several challenging nonlinear test problems are given that illustrate the theory, show how the acceleration improves convergence, show that the quantity predicting the linear convergence rate is remarkably accurate and moreover can be useful for adaptively choosing NGMRES depth, show how restarts can improve convergence in noncontractive iterations, show how NGMRES is naturally suited for finding distinct solutions of a multi-solution PDE, and that NGMRES can perform better than Anderson acceleration when applied to superlinear FPIs.
arXiv:2607.14415v1 Announce Type: new
Abstract: Computed Tomography (CT) carries significant ionizing radiation risks, driving the need for sparse-view reconstruction. Implicit scene representations (ISRs) address this by recovering continuous volumetric attenuation fields directly from sparse projections, and recent geometry-aware extensions jointly model surface geometry alongside attenuation to improve fidelity and enable clean tissue segmentation without manual thresholding. However, these methods remain limited by manually tuned attenuation bounds and rigid two-material constraints. This paper proposes $K$-NeAS, a unified and scalable architecture for automated, multi-material surface reconstruction. We replace independent material networks with a shared latent backbone and introduce a fully differentiable $K$-material sequential soft selector to model an arbitrary number of overlapping tissues. To eliminate manual tuning, we automate attenuation bounding using a Gaussian Mixture Model (GMM) and implement a scheduled auxiliary floater loss to mitigate geometric hallucinations common under extreme sparsity. Evaluated across four clinical Cone-Beam CT (CBCT) datasets, $K$-NeAS successfully scales to arbitrary material counts, achieving superior 3D volumetric fidelity at $K=3$ materials on complex multi-tissue regions such as the Abdomen ($33.28\text{ dB}$ 3D PSNR vs. $31.40\text{ dB}$ single-material NeAS baseline, a $+1.88\text{ dB}$ improvement). Furthermore, our model exhibits enhanced robustness under sparse-sampling conditions, outperforming baseline 3D PSNR by up to $1.17\text{ dB}$ under 5- and 10-view constraints.
arXiv:2607.14314v1 Announce Type: new
Abstract: Seizure diagnosis from EEG signals is a critical yet persistently challenging task, due to the complicated neural dynamics and the spurious connections in inter-channel modeling. While spatial-temporal graph neural networks (STGNNs) have advanced EEG brain network representation learning, the resulting graph structures suffer from low clinical plausibility and limited interpretability due to their purely data-driven nature. To this end, we introduce NeuroGRIP, a retrieval-augmented graph refinement framework that incorporates external medical knowledge to calibrate noisy EEG graphs. We first construct a large-scale, domain-specific knowledge base derived from authoritative clinical guidelines. Leveraging large language models, we extract structured biomedical entities and relations to form a textual knowledge graph (KG), which serves as external knowledge source of clinical priors. Our framework performs alignment-aware query construction by projecting STGNN-generated EEG node embeddings into the semantic space of KG. Semantic queries are then executed via FAISS-based similarity search over knowledge triplets to retrieve relation evidence. Each predicted edge is assigned a confidence score based on retrieved similarity, relation type, and source reliability, enabling us to prune medically implausible edges from the originally predicted graph. Extensive experiments on TUSZ and CHB-MIT demonstrate that NeuroGRIP not only improves seizure detection accuracy but also enhances interpretability by grounding each prediction in clinically validated knowledge. This work provides the first unified framework that tightly couples brain dynamics with external medical expertise via retrieval-augmented reasoning, paving the way for knowledge-enhanced, explainable clinical diagnosis. The code is available at: https://github.com/LincanLi-X/NeuroGRIP.
arXiv:2607.14321v1 Announce Type: new
Abstract: Incorporating hysteresis and eddy currents into finite element simulations of laminated-core electrical machines is computationally challenging. Resolving the fields inside the laminations at each integration point and at every nonlinear iteration leads to computational costs several orders of magnitude higher than anhysteretic simulations, making such approaches impractical for design applications. Conversely, simplified models accounting only for magnetic saturation are becoming increasingly inadequate as electrical machine topologies and operating conditions grow in complexity. In this context, machine learning surrogate modeling has emerged as a promising alternative, offering efficient and accurate approximations of complex electromagnetic behaviors. In this paper, a recurrent neural network is trained as a surrogate of a laminated-core material model for an isotropic laminated core, and is integrated into realistic two-dimensional magnetodynamic finite element simulations based on a magnetic vector potential formulation. The proposed approach achieves excellent agreement with the reference laminated-core model while limiting the computational cost to about twice that of an anhysteretic simulation. By training the recurrent neural network on a sufficiently diverse set of artificially generated magnetic field sequences designed to mimic those encountered in electrical machine simulations, the proposed approach can be readily applied across a wide range of finite element simulations. Furthermore, the trained surrogate model is provided as a standalone component that can be easily incorporated into existing computational frameworks. It is publicly available at https://gitlab.onelab.info/getdp/lamnet.
arXiv:2607.14333v1 Announce Type: new
Abstract: Vision Language Models (VLMs) demonstrate strong perceptual abilities but remain limited in tasks requiring analytical reasoning across multiple visual states, such as multi-image comparison, change detection, and multi-step visual inference. These capabilities are critical for real-world multimodal applications where reasoning must be grounded in systematic differences between visual contexts. However, existing benchmarks rarely require both explicit visual comparison and analytical reasoning, leaving this capability underexplored. To address this gap, we introduce SD-MAR (Synthetic Data for Multi-image Analytical Reasoning), a framework for training and evaluating VLMs on multi-image analytical reasoning. SD-MAR constructs paired visual scenarios through controlled perturbations and generates reasoning tasks spanning semantic change attribution and quantitative comparison. We further train VLMs using GRPO-lite with Backward Discounted Allocation (BDA), a reinforcement learning approach that removes KL regularization to encourage stronger policy optimization while allocating greater credit to the later reasoning steps where analytical conclusions are formed. Experiments on Qwen2.5-VL-7B and InternVL3-8B show that GRPO-lite fine-tuning on SD-MAR improves in-domain accuracy by up to 36.95%, with Qwen2.5-VL-7B outperforming GPT-4.1 on the SD-MAR benchmark. Importantly, out-of-domain generalization is preserved or improved: performance remains within 1% on MME, MMMU-Pro, and MathVista, while improving by up to 4% on MMBench. LLM-as-judge evaluation further demonstrates consistent improvements in logical coherence and explanation quality across both models.