arXiv:2606.17375v1 Announce Type: new Abstract: Boron pebble aggregate was tested for the first time as a high-heat-flux granular plasma-facing material in a tokamak divertor. Exposures of up to $q_{\parallel} = \SI{80}{\MW\per\m\squared}$ incident heat flux were conducted in the DIII-D tokamak. Single protruding rods of pebble aggregate composed of sintered amorphous boron pebbles bound with carbon binder were mounted in the Divertor Material Evaluation System (DiMES) sample holders and exposed to L-mode lower single null (LSN) plasmas. Under these heat loads, significant boron dust emission from the boron spheres was observed, and this dust dominates the divertor boron ionization source. Only about half of the released boron was recovered locally as mm-sized particles; with the rest presumably lost mainly as dust into the plasma and vacuum chamber. Preliminary estimates suggest that the rate of surface recession of $\sim$1 cm/s in the pebble conglomerate within the plasma divertor is consistent with the recession rates observed in laser bench tests subjected to normal-incidence heat loads. Although core performance was not adversely affected by the high boron dust emission, future work will need to improve the boron pebble aggregate design to reduce boron dust emission at high heat loads.
Science Journals
arXiv:2606.17419v1 Announce Type: new Abstract: We develop approximation and generalization error estimates for multi-input neural operators, with the output error measured in Sobolev norms. In contrast to standard operator-learning settings with a single input function, our framework allows multiple input functions defined on possibly different domains, with different dimensions and Sobolev regularities. The derived rates explicitly quantify the contribution of each input space to the final error bound. In particular, in the balanced regime, the approximation and generalization rates are governed by the interaction between the input dimensions, regularities, and Sobolev orders, while the dependence on the model complexity retains a \(\log\log/\log\)-type structure. Our analysis provides a general theoretical framework for multi-input operator learning, including Sobolev training, and is applicable to operator learning problems arising from partial differential equations and scientific computing.
arXiv:2606.17971v1 Announce Type: new Abstract: Parametric PDE-constrained optimal control with pointwise state constraints requires repeated solution of restricted Schur-complement systems on parameter-dependent inactive sets. In a primal active-set method, each inactive-set system is symmetric positive definite, but the active set can change nonsmoothly with the parameter. The resulting operator may vary in dimension, sparsity pattern, and spectrum, limiting reuse of sparse factorizations, multigrid hierarchies, and Krylov information. We propose a reusable spectral-deflation strategy anchored to one full-domain reference Schur complement. Low reference eigenmodes are computed once, restricted online to each inactive set, and used as an A-DEF2 deflation basis for Jacobi-preconditioned CG. The framework also supports POD enrichment, Rayleigh-Ritz reselection, coarse-grid or analytical reference modes, and conditioning safeguards. Given the active set, the method preserves the high-fidelity inactive-set system and solves it to the prescribed CG tolerance; it accelerates the linear algebra rather than replacing the optimal-control solve with a surrogate. We explain the method through a spectral-coherence view, motivated by interlacing and perturbation arguments and assessed with principal-angle diagnostics. Across diffusion, convection-diffusion, nonlinear thermal, and conjugate-heat-transfer benchmarks, deflation reduces CG iterations by about 55 to 98 percent. GPU deployments also show wall-time gains over CPU sparse-direct and algebraic-multigrid baselines, because the reference basis is built once whereas competing solver structures are rebuilt per instance. Coarse-grid or analytical modes amortize the offline cost within a single parameter sweep; fine-grid eigensolves remain more precompute-limited. Timings isolate the inactive-set linear-solve kernel; reducing the active-set outer loop is outside the present scope.
arXiv:2606.17565v1 Announce Type: new Abstract: Massive controlled DC resources (CDCRs), such as battery energy storage systems, are connected to AC power systems through bidirectional inverters for power balance requirements. This study investigates converter-driven stability (CDS) issues in the sub-synchronous frequency range caused by large-scale bidirectional inverter-based stations (IBSs). The impacts of the AC and DC connections of IBSs on subsynchronous oscillations (SSOs) are compared by examining three factors: the number of CDCRs, power flow direction, and control parameters of the inverters. For AC connections, IBSs may induce instability as the number of CDCRs increases, regardless of the power flow direction. To maintain stability, the maximum power amplitude of the IBS is calculated. It is found that switching to DC connections can reduce these instability risks if the DC line resistance is much less than the AC line reactance. Moreover, the method of tuning control parameters is demonstrated to be more effective in improving power-related critical stability under DC connections. Therefore, The DC-IBS is preferred for high-voltage transmission. Finally, the conclusions are validated in power systems connected with both AC- and DC-IBSs under various network topologies and system scales.
arXiv:2603.18492v3 Announce Type: replace Abstract: Mixture-of-Experts (MoE) language models increase parameter capacity without proportional per-token computation, yet deployment still requires storing the full expert pool, making expert pruning important for reducing memory and serving overhead. Existing task-agnostic expert-pruning methods are typically calibration-dependent: they estimate expert importance from routing or activation statistics on a calibration set, making pruning decisions sensitive to calibration-data variation while introducing substantial preprocessing cost. We propose AIMER (\textbf{A}bsolute mean over root mean square \textbf{IM}portance for \textbf{E}xpert \textbf{R}anking), a simple calibration-free criterion that identifies more distinct experts by capturing the concentration pattern of expert weights, making it well suited for task-agnostic expert pruning. Across 7B to 47B MoE language models with distinct architectures and 16 diverse benchmarks, AIMER consistently delivers stronger capability balance across diverse tasks than existing calibration-free methods. Surprisingly, AIMER also achieves better balance than strong calibration-based expert-pruning baselines calibrated on the widely used task-agnostic C4 corpus, while requiring only 0.22--2.06 seconds to score all experts.
arXiv:2606.00024v3 Announce Type: replace Abstract: Long-context decoding in Large Language Models (LLMs) is constrained by the cost of accessing and processing the Key-Value (KV) cache. Despite evidence that attention outputs depend jointly on keys and values, most existing KV management methods rely on key-only pruning, since incorporating values incurs prohibitive overhead. In this paper, we propose Attention Run-time Termination (ART), a lightweight run-time mechanism that tracks accumulated attention outputs during kernel execution and terminates subsequent KV block accesses once further contributions become negligible. Rather than replacing KV selection, ART dynamically terminates redundant KV traversal on top of existing dense or sparse attention policies. We introduce a stability-based criterion that monitors both magnitude and directional changes of intermediate attention outputs and provideds a theoretical characterization of the resulting truncation error. Experiments on the LongBench and RULER Needle-in-a-Haystack tasks show that ART increases the generation throughput of existing KV-cache methods by up to 20%, without compromising the result quality.
arXiv:2606.18108v1 Announce Type: cross Abstract: We develop a text-to-SQL (structured query language) system based on large language models (LLMs) using in-context learning and apply it to the Automatic Learning for the Rapid Classification of Events (ALeRCE) astronomical database. ALeRCE is a community broker for the Zwicky Transient Facility and the Vera C. Rubin Observatory. The system enables users to query the database in natural language (NL) and generates executable SQL queries. To develop and evaluate the system, we constructed a dataset of 110 NL/SQL pairs. We propose a step-by-step generation framework comprising four modules: schema linking, query classification, prompt decomposition, and self-correction. The performance of thirteen LLMs is evaluated using in-context learning and prompt engineering techniques. Text-to-SQL performance is assessed using the perfect-match (PM) rate for row identifiers (e.g., object identifiers) and column identifiers (i.e., column names). The proposed step-by-step framework consistently outperforms a direct-inference baseline, while the self-correction module consistently reduces execution errors. For Claude Opus 4.6, PM performance on row (column) identifiers is high for simple queries, reaching 0.97 (0.94), and decreases with query complexity to 0.44 (0.72) for medium queries and 0.59 (0.49) for hard queries. Among the thirteen evaluated models, the best-performing LLMs for the text-to-SQL task are Claude Opus 4.6, Gemini 2.5 Pro, Gemini 3 Flash, and GPT-5.2-Codex.
arXiv:2606.18150v1 Announce Type: cross Abstract: Channel state information (CSI)-based neural positioning learns a mapping from CSI measurements to user equipment (UE) positions using neural networks. However, most existing performance evaluations utilize randomly partitioned train/test CSI-dataset splits, which fail to reflect the generalization requirements of practical deployments and present optimistic results. In this paper, we study the spatial and temporal generalization of neural positioning with standard-compliant Wi-Fi and 5G NR systems for three real-world CSI datasets acquired in indoor and outdoor environments. We assess generalization with two different architectures, a conventional multilayer perceptron (MLP) and a novel transformer architecture, to unseen spatial regions, unseen UE trajectories, and CSI measurement campaigns separated by one week. Our experiments show that both architectures generalize well in space and time, and the proposed transformer consistently outperforms the MLP in positioning accuracy while requiring fewer model parameters.
arXiv:2606.17399v1 Announce Type: new Abstract: When small transformers grok modular multiplication, prior work reports that the learned embedding has a "dense" Fourier spectrum requiring all frequencies. This contrasts with modular addition, where only a sparse set of key frequencies suffices. We show this density is an artifact of analyzing in the wrong basis. The natural Fourier transform for multiplication is not the standard additive DFT but the multiplicative character transform, which decomposes functions on the multiplicative group $(\mathbb{Z}/p\mathbb{Z})^*$ into its irreducible representations. Applying this transform to a grokked transformer trained on $a \cdot b \bmod 113$, we find the embedding spectrum becomes highly sparse (Gini coefficient 0.58 vs. 0.07 in the additive basis) with only 4 key frequencies carrying significant energy. Furthermore, 96.9% of MLP neurons are cleanly tuned to a single multiplicative frequency, and neuron activation heatmaps reveal 2D-periodic structure when reordered by the discrete logarithm. These results demonstrate the transformer reduces multiplication to addition in discrete-log space, implementing a "Discrete-Log Clock" algorithm analogous to Nanda et al.'s Clock algorithm for addition. The methodology generalizes: matching the analysis basis to the algebraic structure of the task reveals interpretable structure where standard tools see noise.
arXiv:2606.17071v1 Announce Type: new Abstract: Einstein field equations allow cosmological dynamics to depart from the Friedmann-Lemaitre-Robertson-Walker (FLRW) idealisation in several physically different ways. Matter may become spatially inhomogeneous, the local expansion scalar may vary across a hypersurface, the expansion may acquire anisotropic components through shear, and the free gravitational field may be encoded in nonzero Weyl curvature. The key question is not only how far a model is from FLRW, but which geometric mechanism is responsible. A single departure from FLRW number cannot distinguish these mechanisms. This paper introduces a compact geometric diagnostic framework that keeps them separate while using standard quantities in general relativity. The framework is observer-explicit and domain-explicit, intended as a practical tool for comparing analytic and numerical solution families rather than as a new invariant classification of spacetime. Buchert's kinematical backreaction is retained as a derived explanatory quantity rather than a separate axis, since it is already fixed by the expansion-variance and shear contributions. A single curvature normalisation is used for all Weyl diagnostics. The method is tested on six benchmarks, namely FLRW, Bianchi-I, Kasner, Lemaitre-Tolman-Bondi dust, scalar-perturbed FLRW, and tensor-perturbed FLRW. These benchmarks occupy distinct regions of the diagnostic space, and the magnetic Weyl contribution appears only in the tensor case. The classification remains stable under changes of perturbation amplitude, spatial resolution, averaging domain, constraint reliability, and a leading-order observer tilt. The curvature expressions for the exact benchmarks are verified symbolically against metric-derived Weyl invariants, and the supporting computer code, numerical results, tables, and figures are publicly available.
arXiv:2606.17405v1 Announce Type: new Abstract: Clinical decision support AI systems (CDSASs) must adapt to evolving patient conditions in real-time while adhering to strict safety constraints. We present an online adaptive framework that integrates Treatment Effect (TE) estimation to quantify clinical benefits, a patient Digital Twin (DT) to simulate treatment trajectories, and Reinforcement Learning (RL) for sequential decision-making. The AI system is initially trained on historical medical records and operates in a continuous learning loop. To ensure safety, a rule-based module monitors vital signs and blocks contraindicated treatments. Cases with strong internal model disagreement are flagged for clinician review, simulated in our experiments via a pre-trained outcome model. We validate our framework using both a synthetic clinical simulator and a real-world ovarian cancer dataset from The Cancer Genome Atlas (TCGA). In both simulated and clinical settings, our method demonstrated superior effectiveness and stability in recommending treatments compared to standard computational baselines. Furthermore, the AI system maintains low latency and requires expert consultation for only a minority of cases in our experimental validation, demonstrating its potential as a safe, clinician-supervised tool for personalized medicine that continuously improves through practical use.
arXiv:2606.17376v1 Announce Type: new Abstract: Respiratory-rate (RR) monitoring is a critical component of remote triage and victim assessment in emergency response, disaster recovery, and infectious-disease scenarios, where minimizing physical contact can reduce responder risk and improve operational safety. However, field deployment of contactless RR monitoring remains challenging due to variable illumination, posture changes, platform heterogeneity, and the impracticality of wearable sensors in hazardous environments. In this paper, we present a modality-adaptive contactless RR monitoring framework for heterogeneous mobile robots with onboard edge computing. The proposed system combines brightness-adaptive sensor selection across RGB, thermal, near-infrared (NIR), and low-light cameras, keypoint-guided chest ROI extraction for posture-robust monitoring, and a signal-quality-index (SQI)-based filtering mechanism for reliable respiratory estimation. We implement and evaluate the framework on three robotic platforms spanning quadruped and wheeled locomotion and multiple edge-computing architectures. Experiments conducted across diverse lighting conditions, subject poses, and robot-to-subject distances demonstrate that the framework generalizes across platforms without per-platform algorithmic retuning, while revealing modality-specific operational boundaries. RGB provides the broadest coverage up to 8m, NIR remains effective up to 6m, thermal is reliable only at short range, and low-light sensing supports monitoring in complete darkness up to 8m. Overall, the results demonstrate the feasibility of multimodal contactless RR monitoring on mobile robots and support its use as a foundation for autonomous triage and victim assessment in hazardous search-and-rescue settings.
arXiv:2606.17387v1 Announce Type: new Abstract: In recent decades, privacy-enhancing technologies (PETs) have been recognized as a means of meeting regulatory and user privacy requirements in software systems that process personal data. Despite substantial research efforts, support from regulators, contributions by large technology companies such as Google and Microsoft, and growing interest among software practitioners, the practical adoption of PETs remains limited. Existing research consistently identifies recurring challenges to PETs adoption in SE, such as technical complexity and insufficient training. Despite ongoing research efforts, these challenges largely remain unresolved in practice. In this industrial challenge paper, we apply a practical, requirements engineering (RE)-driven perspective to examine challenges to PET adoption across multiple stakeholder groups (PET developers, integrators, and adopters) as well as across different disciplinary perspectives (engineering, law, and business). We argue that RE can facilitate the adoption of PETs by systematically addressing each of the complementary engineering, business, and legal viewpoints on privacy. Neglecting challenges in any of these viewpoints (e.g., the impact of PETs on software architecture, their business implications, and their contribution to regulatory compliance) can increase the impediments or even lead to implementation failure. In practice, explicit specification of these viewpoints within RE can enable meaningful coordination among stakeholders to more effectively realize the benefits of PETs in software engineering.
arXiv:2606.17588v1 Announce Type: new Abstract: Several studies have examined the use of large language models (LLMs) for title-abstract screening in systematic reviews (SRs), reporting mixed accuracy. However, questions of reliability remain largely unaddressed. In this study, we go beyond quantitative LLM-human agreement metrics and qualitatively investigate how and why LLMs fail. We also propose actionable recommendations. We analyzed disagreements between LLMs and researchers across six software engineering SRs and over 1,000 primary study papers. For each SR, papers were screened independently by human experts and LLMs in zero-shot mode, resulting in Kappa values ranging from 0.52 to 0.77. Qualitative analysis suggests that human-LLM disagreement results from recurring, identifiable causes, such as boundary ambiguity in key terms, keyword overemphasization, and incorrect topic inference. Based on these findings, we propose recommendations such as validating semantic understanding before deployment, running multiple LLMs, and focusing validation efforts on borderline cases. Future studies are needed to validate the impact of our recommendations, and community efforts are needed to develop normative guidelines on LLM usage in SRs.
arXiv:2606.17078v1 Announce Type: new Abstract: Binary collision approximation (BCA) codes are potentially powerful tools to simulate ion irradiation ejecta properties, such as the composition and the angular and energy distributions of the sputter yield. However, recent advances in the sputtering of minerals have highlighted the low predictive fidelity of BCA codes such as SDTrimSP when compared to experimental measurements. We demonstrate how a sputtering model that underestimates the forward sputtering on a flat surface at large ion incidence angles from surface normal will lead to an erroneous result for rough and porous surfaces, where most ejected particles are directed along the surface normal. We demonstrate how this is the case for an existing model, which reliably predicts sputtering mass yields from a flat enstatite surface but fails to accurately reproduce the angular distribution of sputtered particles. We then compare this to a BCA model incorporating higher surface-binding energies$-$based on a molecular dynamics description of plagioclase$-$which underestimates mass yields but significantly reduces back-sputtering and better reproduces laboratory sputter angle distributions measured at large ion incidence angles. We conclude that the BCA model cannot simultaneously reproduce both the sputter yield and the sputter angle distribution arising from He irradiation of mineral targets, either due to the inherent geometric simplicity of the BCA or because the model neglects yield-enhancing processes such as molecule and cluster sputtering. This demonstrates a structural limitation of current BCA-based models when realistic surface morphologies are considered, rather than a problem that can be resolved by parameter tuning alone.
arXiv:2606.16070v2 Announce Type: replace Abstract: World-model synthesis aims to turn interaction experience into an internal model of environment dynamics. Existing symbolic approaches often fit observed transitions or mixtures of local rules, but they do not produce a complete executable program that can run independently of the real environment. We present Mind-Studio, a framework that synthesizes executable pygame-style world models from state-action-next-state trajectories using large language models. Mind-Studio combines entropy-selected traces with a lightweight game skill file containing object, action, and static scene information extracted from screenshots. We evaluate synthesis quality with a K-step lookahead fidelity protocol that compares generated world-model rollouts against Real-ALE rollouts from the same state. On Montezuma's Revenge, Mind-Studio improves chosen-action next-state prediction from 0.3% for PoE-World to 48.7% while verifying 5 of 8 subgoals; across Alien, Assault, and Skiing, it achieves stronger branch-level fidelity than prior learned lookahead sources.
arXiv:2505.19222v3 Announce Type: replace Abstract: We are concerned with discretisations of the classical Kolmogorov equation by a standard space-time discontinuous Galerkin method. {The} Kolmogorov equation serves as simple, yet rich enough in the present context, model problem for a wide range of kinetic-type equations: although it involves diffusion in one of the two spatial dimensions only, the combined nature of the first order transport/drift term and the degenerate diffusion are sufficient to `propagate dissipation' across the spatial domain in its entirety. This is a manifestation of the celebrated concept of hypocoercivity, a term coined and studied extensively by Villani in \cite{villani}. We show that the {classical} space-time discontinuous Galerkin method {admits} a corresponding hypocoercivity property at the discrete level, asymptotically for large times. To the best of our knowledge, this is the first result of this kind for any standard Galerkin scheme. This property is shown by proving one part of a discrete inf-sup-type stability result for the method in a family of norms dictated by a modified scalar product motivated by the theory in \cite{villani}. This family of norms contains the full gradient of the numerical solution, thereby allowing for a full spectral gap/Poincar\'e-type inequality at the discrete level, thus, showcasing a subtle, discretisation-parameter-dependent, numerical hypocoercivity property. Further, we show that the space-time discontinuous Galerkin method is inf-sup stable in the family of norms containing the full gradient of the numerical solution, which may be a result of independent interest.
arXiv:2505.19937v3 Announce Type: replace Abstract: Large Language Models (LLMs) are extended into Speech-LLMs, and the quality of the audio--text alignment they learn affects most downstream Spoken Language Understanding (SLU) behavior. Yet despite a growth of fusion strategies, there is no standard way to measure how well a Speech-LLM internally binds audio frames to text tokens. We introduce ALAS (Automatic Latent Alignment Score), a model and task-agnostic metric that probes the LLM's per-layer hidden states, scoring the cross-modal cosine similarity between audio and text representations against a Whisper-derived reference. ALAS needs only a frozen forward pass and an off-the-shelf ASR reference, with no training or fitted classifier, and is calibrated to an interpretable uniform baseline comparable across tasks. Applying ALAS to four open-source Speech-LLMs (AF3, Qwen2-Audio, Qwen-Omni, SALMONN) across emotion recognition (IEMOCAP), open-ended SQA (LibriSQA), and multi-choice audio understanding (MMAU-speech), we find that the depth and strength of alignment reflect each model's audio-encoder design and the acoustic-versus-semantic demands of the task, and that ALAS tracks but does not duplicate task accuracy, exposing models that score well without genuinely grounding in the audio. We release ALAS as an open-source library so that practitioners can probe their own Speech-LLMs or try it on new tasks.
arXiv:2606.18153v1 Announce Type: new Abstract: The Open Forest Observatory (OFO) is a collaboration across universities and other partners to make low-cost forest mapping accessible to ecologists, land managers, and the general public. The OFO is building both a database of geospatial forest data as well as open-source methods and tools for forest mapping by uncrewed aerial vehicle. Such data are useful for a variety of climate applications including prioritizing reforestation efforts, informing wildfire hazard reduction, and monitoring carbon sequestration. In the current iteration of the OFO's forest map database, 3D tree maps are created using classical structure-from-motion techniques. This approach is prone to artifacts, lacks detail, and has particular difficulty on the forest floor where the input data (overhead imagery) has limited visibility. These reconstruction errors can potentially propagate to the downstream scientific tasks (e.g. a wildfire simulation.) Advances in 3D reconstruction, including methods like Neural Radiance Fields (NeRF), produce higher quality results that are more robust to sparse views and support data-driven priors. We explore ways to incorporate NeRFs into the OFO dataset, outline future work to support even more state-of-the-art 3D vision models, and describe the importance of high-quality 3D reconstructions for forestry applications.
arXiv:2606.18154v1 Announce Type: new Abstract: Building personalized cardiac electrophysiology (EP) digital twins requires identifying the appropriate model structure for each patient, not merely fitting parameters. Traditional methods rely on experts to manually prescribe hybrid physics-neural architectures, which requires deep domain expertise and does not transfer across patients. Recent works have applied large language models (LLMs) to generate or act as hybrid models. However, despite their promising generalization capacity, these LLM-based methods lack the structural priors needed for stable cardiac simulations. Hence, we propose LEADS, a framework that formulates cardiac EP domain knowledge as a structured action space and utilizes an LLM agent to discover hybrid models. The agent follows an iterative reasoning-and-action loop to select, combine, and refine hybrid models, whilst gradient descent handles parameter fitting. The proposed LEADS designs every candidate model towards physically grounded, interpretable, and numerically stable, while allowing open-ended architectural discovery. We validate LEADS on synthetic data with three ground-truth reaction models and on real cardiac EP data, demonstrating that it outperforms both human-designed hybrid models and other LLM-based hybrid modeling.
arXiv:2606.17398v1 Announce Type: new Abstract: Binary reversing is fundamental to software understanding, vulnerability discovery, malware investigation, and firmware auditing. However, it remains inherently challenging due to the irreversible loss of semantic information during compilation. Recent advances in machine learning, large language models (LLMs), and agentic AI systems have accelerated the adoption of AI-augmented binary reversing. Yet, the resulting body of work has become increasingly fragmented across reversing domains, artifact representations, learning approaches, and evaluation practices. This paper presents the first comprehensive systematization of knowledge on AI-augmented binary reversing. We analyze 144 research papers published since 2015, and organize them into 22 binary reversing domains according to the inference tasks. We further introduce a unified taxonomy spanning conventional and AI-augmented reversing pipelines. Our taxonomy connects traditional analysis techniques, binary-derived artifacts, representation strategies, learning paradigms, and downstream inference tasks, while clarifying the emerging roles of LLMs and agentic AI systems. By establishing a common vocabulary and structured framework, we provide a holistic view of the field's evolution over the past decade. Our study reveals common structures underlying seemingly disparate approaches, highlights persistent technical challenges and evaluation gaps, and identifies promising opportunities for future research. Collectively, these insights clarify the current state of the field and provide a foundation for the next generation of reliable and scalable AI-augmented binary reversing systems.
arXiv:2606.17406v1 Announce Type: new Abstract: Feature extraction involves the identification and extraction of salient characteristics or patterns, including edges, textures, shapes, and color attributes. Contemporary feature extractors predominantly leverage deep learning architectures, such as Convolutional Neural Networks (CNNs) and Vision Transformers (VITs). The availability of diverse feature extractors in the literature provides a wide range of feature representations. Features extracted from an image depend on the specific application, the chosen extractor, and its configuration. Therefore, integrating complementary information by combining distinct extractors offers a promising way to enhance performance. Graph Neural Networks (GNNs), particularly Graph Convolutional Networks (GCNs), have emerged as powerful and widely adopted approaches for semi-supervised image classification, as they effectively leverage both labeled and unlabeled data while exploiting the underlying graph structures that capture relationships among samples. This study proposes a novel approach for GNNs in scenarios where labeled data is scarce, by integrating diverse sets of feature and graph representations derived from various extractors in classification scenarios. Experimental investigations were conducted, encompassing combinations of distinct feature and graph extractors, as well as rank aggregation strategies. The primary contributions of this work are underscored by the experimental findings, which demonstrate that the strategic combination of feature and graph representations, coupled with the application of manifold learning for graph processing, leads to significant improvements in classification accuracy across the majority of experimental conditions. Furthermore, the utilization of rank aggregation techniques to integrate features from different extractors was shown to enhance classification accuracy.
arXiv:2606.17123v1 Announce Type: new Abstract: In open large language model (LLM) ecosystems, models are frequently adapted across multiple domains and applications, forming multi-stage derivation chains. Consequently, tracking and verifying historical contributions is essential for model provenance and intellectual property protection. However, existing watermarking methods are mainly designed for single-user, one-time embeddings, often fail under repeated model derivation and incremental updates. To address this problem, we propose LineageMark, a multi-user white-box watermarking framework for model derivation chains. The framework encodes watermarks in model parameters using a projection-based approach. Stable carriers are first selected to reduce sensitivity to model changes, each watermark bit is then represented as a projection statistic over these carriers. Additional watermark insertions introduce only bounded perturbations in the projection space, and margin constraints are used to maintain signal integrity. We evaluate the effectiveness of LineageMark in multi-stage model derivation chains. Experimental results show that LineageMark preserves contributor watermarks across multi-stage derivation and supports incremental multi-user watermark insertion. Furthermore, it exhibits robustness against perturbations such as re-watermarking, fine-tuning, quantization, and pruning.
arXiv:2606.17183v1 Announce Type: new Abstract: Answering navigation-relevant questions over long egocentric videos requires retrieving and organizing evidence distributed across distant temporal moments while maintaining spatial and contextual consistency. Although long-context vision--language models can achieve strong answer quality, they are computationally expensive for long trajectories and inefficient for repeated querying. Recent graph-based approaches such as VL-KnG address this challenge through persistent spatio-temporal knowledge graphs, but graph-centric retrieval alone may underrepresent broader temporal continuity and contextual cues. We present VL-MemKnG, a hybrid memory framework that extends VL-KnG by combining a spatio-temporal knowledge graph with persistent segment-level contextual memory. The knowledge graph captures structured relational information and long-range object associations, while segment-level memory preserves broader temporal context for long-horizon evidence retrieval. A hybrid retrieval-and-reasoning module jointly operates over both memory representations to produce evidence-grounded answers and temporally organized supporting evidence. We also introduce WalkieKnowledgeT+, an extension of WalkieKnowledge for long-horizon navigation-oriented video question answering. The benchmark includes temporally distributed reasoning tasks requiring evidence aggregation across multiple non-cooccurring moments. On WalkieKnowledgeT+, VL-MemKnG improves Top-1 retrieval accuracy from 58% to 67% and Recall@1 from 34.50% to 40.55%, outperforming all compared methods, including Gemini 2.5 Pro and Qwen 3.5+. The gains are particularly pronounced on temporal-global and temporally scattered aggregation questions, demonstrating the benefits of combining structured relational memory with segment-level contextual memory while maintaining efficient query-time inference.
arXiv:2606.17212v1 Announce Type: new Abstract: Dynamic Gaussian compression is normally optimized for complete files or complete progressive prefixes, but interactive rendering encounters partial representations: some spatiotemporal regions are present, others missing, and late refinements cannot affect the displayed frame. We study dynamic Gaussian representations whose incomplete delivery states remain directly renderable and whose degradation is optimized in image space. Gaussian primitives are organized into independently addressable spatiotemporal clusters with a base level and three refinements; training samples partial dependency graphs, renders many counterfactual states in one GPU batch, and minimizes expected distortion, tail distortion, temporal inconsistency, rate, and prefix regressions. A counterfactual utility layer measures the marginal render contribution of each completion group across valid receiver contexts. The same graph admits a concrete delivery realization with MTU-bounded entropy-coded chunks, deadline-aware scheduling, and receiver-side dependency closure. On held-out views, the finest refinement has negative mean marginal utility in 3/32 D-NeRF bouncingballs, 49/64 HyperNeRF broom2, and 28/64 HyperNeRF chicken clusters; its lower-tail utility is negative in 21/32, 61/64, and 42/64 clusters, respectively. On broom2, render-utility ordering removes both PSNR regressions produced by nominal layer order at matched byte budgets; on chicken, utilities measured on disjoint training cameras improve held-out PSNR by 3.03 dB at the lowest matched budget. These scoped results show why nominal refinement order cannot substitute for render-conditioned utility: the formulation treats network delivery as a distribution over renderable scene states rather than as an external wrapper around a graphics codec.