arXiv:2607.10822v1 Announce Type: new
Abstract: Indoor spatial understanding remains a fundamental challenge for intelligent systems operating in physical environments. Traditional RFID localization techniques typically estimate positions of tags using signal strength measurements but fail to capture higher-order spatial relationships between objects and infrastructure. Recent work on RFID and wireless indoor localization has increasingly emphasized robust learning under noisy propagation, while recent graph-based localization methods demonstrate the value of relational modeling over isolated samples. This paper introduces a graph-based learning framework that leverages Graph Neural Networks (GNNs) to infer spatial geometry from RFID observations. Rather than predicting isolated coordinates, the proposed system models relationships between RFID readings, antennas, and physical structures within an indoor floorplan. This framing is aligned with recent graph-based indoor positioning and graph construction literature, where topology is a first-class source of information for downstream inference. The approach integrates signal strength data, floorplan semantics, and spatial constraints into a graph representation where nodes correspond to RFID observations and edges encode proximity and contextual relationships. A GNN is then trained to predict geometric patterns such as linear trajectories, rectangular bounding regions, and movement paths of objects in space.
Science Journals
arXiv:2607.10885v1 Announce Type: new
Abstract: The dual lattices $E_6^*$ and $E_7^*$ are of particular interest in source coding and data compression applications. Among all known lattices in dimensions six and seven they attain the smallest normalized second moments, i.e., the smallest average quantization error. Their use in practice requires fast closest-point (nearest-lattice-point) algorithms. The known approach, due to Conway and Sloane and completed for $E_6$ and $E_6^*$ by Takizawa, Yagi, and Kawabata (TYK), decodes these lattices as unions of cosets of root lattices $A_n$: each coset is decoded separately, and the best result is kept. This requires four coset decodings for $E_7^*$ and six for $E_6^*$, together with explicit distance computations.
This paper shows that all these coset decodings can be collapsed into a single sweep. Reformulated in terms of glue vectors, the TYK decompositions state that $E_7^*$ is the union of the even glue classes of $A_7^*$, and that $E_6^*$ is a parity-matched sublattice of $A_1^*\oplus A_5^*$. The candidate chain constructed by the closest-point algorithm of McKilliam, Clarkson, and Quinn (MCQ) for $A_n^*$ visits every glue class of $A_n$ exactly once and is optimal within each class. Consequently, one sorted sweep per coordinate block yields the closest points of all glue cosets simultaneously, and $E_6^*$ and $E_7^*$ are decoded at roughly the cost of a single $A_5^*$ or $A_7^*$ quantization. Rough operation counts indicate a $4$--$6\times$ reduction for $E_6^*$ and $3$--$4\times$ for $E_7^*$ relative to coset-by-coset decoding. We also discuss further constant-factor improvements available from recent refinements of the $A_n^*$ algorithms, and an open question concerning sort-free linear-time decoding.
arXiv:2607.11052v1 Announce Type: new
Abstract: Machine learning progress is often attributed to scaling model size and dataset volume, yet the composition of data can be just as consequential. Empirical findings repeatedly show that combining datasets from different domains yields nontrivial interactions. For instance, adding code improves mathematical reasoning, while certain mixtures introduce interference that reduces model performance. We refer to these effects collectively as data synergy, where the contribution of multiple domains exceeds or falls short of the sum of their isolated contributions. In this work, we formalize and quantify data synergy in language model pretraining. Leveraging observational variation across open-weight LLMs with diverse pretraining mixtures, we estimate both direct domain-to-benchmark synergy (how one domain contributes to performance on another) and a second-order domain-domain synergy (capabilities that require co-occurrence of multiple domains). Our framework improves predictive accuracy over domain-agnostic scaling laws and recovers stable synergy estimates. We validate these estimates by training models on predicted optimal and predicted anti-optimal mixtures and confirm that our synergy estimates correctly predict performance rankings.
arXiv:2607.11680v1 Announce Type: new
Abstract: Prompt engineering is commonly presented as a technical competence for obtaining more accurate, relevant, or well-formatted outputs from large language models (LLMs). However, in STEM education, prompting should also be understood as a continuous epistemic practice. Students interpret contextual and disciplinary cues and adopt expectations about what kind of knowledge, representation, and action are appropriate. Drawing on epistemological framing, and AI-mediated concept-to-decision reasoning, the paper presents a new framework called epistemic prompting and proposes a multi-turn Framing-Prompting Loop. The educationally relevant outcome is a prompt framing trajectory: the sequence of prompts, model responses, learner uptake, disciplinary checks, and reframing moves through which a knowledge task develops. In this framework, an initial prompt establishes a provisional macro-frame by selecting the problem, representations, assumptions, criteria, and distribution of work between learner and model. Each subsequent learner turn can then maintain, specify, challenge, repair, or transform that organization. The implications for AI-mediated STEM instruction, and, specifically, on learner-LLM interaction are also discussed.
arXiv:2607.10925v1 Announce Type: new
Abstract: This paper presents a framework for multi-session mapping of underwater environments utilizing an affordable action camera. The Visual-Inertial data are augmented by water depth recordings from a dive computer. SVIn2, an open-source VI-SLAM framework, is utilized to generate a trajectory and a sparse reconstruction for each session. Utilizing the keyframes extracted from SVIn2 and the estimated camera poses, a Structure-from-Motion (SfM) framework, COLMAP, is employed for global optimization and to produce a dense reconstruction of the target environment. The presence of calibration targets at fixed locations, when available, is used to estimate the coordinate transformation between different data collection sessions, thus transforming the different sessions into the same coordinate frame. The proposed pipeline is employed for the mapping of a shipwreck off the coast of Barbados. For the first time, both the exterior and the accessible interior parts of the wreck were mapped in two sessions, while a third session employed two cameras with different fields of view.
arXiv:2607.10930v1 Announce Type: new
Abstract: Generative models often represent signals as dense grids of amplitudes, blurring sharp transients that are crucial for the correctness of physical signals. We introduce Singularity Space, a generative framework that represents signals through complex-plane singularities, rooted in the classical pole-residue representation of meromorphic functions. We learn a latent space of physically constrained, per-signal singularity configurations to solve an inverse problem from degraded or partial observations. The framework has three key properties: interpretability, in which each generated singularity configuration corresponds to a set of physical parameters; structural stability, which mitigates Gibbs artifacts at discontinuities; and resolution-free output reconstruction on arbitrary grids without retraining or interpolation. Our framework employs a transformer-based diffusion model that directly predicts samples at complex-plane singularity coordinates, subject to geometric constraints during sampling. As a controlled test case for sharp-feature recovery, we evaluate our framework on 1D Burgers shocks, where each shock is represented by 32 predicted singularities (an $8\times$ reduction versus a 1024-point grid signal). Our framework preserves signal structure ($\text{TV ratio} \approx 1$) under unseen test-time observation noise, achieves a $4.2\times$ lower reconstruction error in zero-shot sub-resolution generalization than a grid-based baseline, and recovers physical parameters to $10^{-4}$ absolute error in-distribution. These results suggest that singularity-based representations may provide a practical foundation for other transient-dominated signals such as speech and biomedical signals, with potential extension to higher-dimensional domains.
arXiv:2607.11226v1 Announce Type: new
Abstract: LLM agents today are caught in an awkward bind. Lock them down with static safety instructions and they rarely venture beyond the obvious; give them free reign with tools and multi-agent debate, and safety violations quickly follow. Rather than forcing a single model to juggle both creativity and caution, we separate the concerns across specialized roles. A Disrupter generates unconventional proposals, a Validator enforces hard runtime checks at the tool gateway, and a Broker pulls in distant but relevant analogies. Failures are not discarded -- they are compiled, via MCTS, into compact, signed constraint patches we call Scars. These patches are cached locally and inherited by future cohorts, turning repeated failures into reusable, low-cost runtime constraints. In a spatial-semantic sandbox (N=20 runs, p<0.01), our cohort reaches remote targets where debate fails, the Validator prevents all executed breaches, and Scars reduce token consumption by 15.1% by avoiding redundant validator checks. Furthermore, credit-based Communication Allocation Scores (CAS) restrict outbound bandwidth, reducing overall token costs by 55.9% under resource constraints.
arXiv:2607.11073v1 Announce Type: new
Abstract: Accurately representing atmospheric aerosol populations is essential for simulating aerosol-cloud interactions, radiative forcing, and ice nucleation, yet existing reduced schemes impose structural assumptions that limit their ability to capture composition diversity and mixing state. Machine-learning approaches offer more flexible representations, but standard autoencoders do not preserve the mathematical structure of aerosol populations and therefore cannot support physically meaningful process operators. We introduce AeroMELD (Aerosol Measure Embedding for Latent Dynamics), a mathematically grounded framework for constructing low-dimensional latent variables that retain this structure. We show that any permutation-invariant linear encoder must take a scale-shape decomposition, with total number concentration represented explicitly and latent shape given by a barycentric combination of per-particle embeddings. This aggregated latent state retains the diagnostic expressiveness of a Deep Sets model by moving the nonlinear post-aggregation stage into the learned diagnostic map while preserving latent linearity. Using particle-resolved data as ground truth, we encode weighted particle populations directly rather than binned aerosol states; size-resolved mass and number distributions serve only as diagnostic targets and visual summaries. The latent space accurately reconstructs these distributions, CCN spectra, optical coefficients, and immersion-freezing behavior while preserving the linear population structure needed for hybrid ML-physics models. Although the experiments focus on diagnostic reconstruction, the embedding is designed so that emissions and mixing can be represented exactly and nonlinear microphysical processes learned in a controlled latent space. This work establishes a foundation for learning aerosol-process evolution directly in latent space.
arXiv:2607.10962v1 Announce Type: new
Abstract: In this paper, a multidisciplinary design optimization problem under uncertainty is formulated for wave energy converter array. An array of heaving point absorbers for grid-scale energy production with decision variables and parameters chosen from the coupled disciplines of geometry, hydrodynamics, layout, and trajectory optimization thus resulting in a control co-design formulation of the plant and the control together. We study the benefits of MDO as applied to WEC farm layout optimization. We vary the wave energy converter (WEC) dimensions, array layout, and control gain to minimize the power per volume. Uncertainty in the electrical power is handled using regression based on polynomial chaos expansion (PCE) method at each design iteration. Traditional WEC farm design optimization approaches often neglect the multidisciplinary, coupled nature of WECs and the inherent uncertainty in ocean wave conditions and control responses. This leads to designs that may under perform in real-world environments. In this work, we address this limitation by incorporating uncertainty directly into the design optimization process using the technique of polynomial chaos expansion (PCE) to quantify the variability of the performance due to uncertain wave environment.
arXiv:2607.09708v1 Announce Type: new
Abstract: Solving large-scale instances of the Traveling Salesman Problem (TSP) exactly is computationally expensive. Researchers often employ graph sparsification methods to improve computational efficiency. Traditional sparsification methods typically rely on fixed heuristics and fail to fully exploit instance-specific structural information. In this paper, we propose Graph Edge Sparsification (GES), a learning-based sparsification approach for Euclidean TSP. By incorporating geometric structural information and combinatorial optimization technology, our proposed method adaptively generates a sparsification graph for different instances, significantly reducing the graph size and accelerating the solving process. Experimental results demonstrate that our sparsification method can prune up to 95% of edges on the MATILDA dataset, while keeping the solution gap within 1% of the optimal value. Moreover, our approach exhibits strong generalization capability on the TSPLIB benchmark.In some large-scale instances, the pruning rate exceeds 99%, while the optimality gap remains below 1%.
arXiv:2607.11071v1 Announce Type: new
Abstract: Single-image dehazing aims to recover clear scenes from haze-degraded images. It remains challenging due to the atmospheric scattering and the complexity of real-world haze distributions. Although recent end-to-end networks have achieved promising performance, two issues still limit their effectiveness: insufficient feature refinement at the bottleneck stage and weak local structural representation in encoder-decoder architectures. Thus, we propose a Haze-Aware Dual-Domain Refinement Network (DDR-Net) for single-image dehazing. Our method is built upon three modules: Haze Prior Extractor (HPE) provides multi-scale haze-aware priors by operating directly on downsampled hazy images; Detail-Enhanced Blocks (DE Blocks) serve as the core feature extraction units, capturing multi-scale structural information and enhancing edge and texture recovery via gradient-aware convolutions; and Spatial-Frequency Bottleneck Refinement (SFBR) at the bottleneck jointly exploits spatial and frequency information to refine bottleneck features. DDR-Net achieves more effective feature representation and reconstruction for haze removal. Extensive experiments on real-world benchmarks demonstrate that our method outperforms existing dehazing approaches. It achieves competitive performance on synthetic datasets.
arXiv:2607.11703v1 Announce Type: new
Abstract: The asymmetry between language production and perception has been well-documented in psycholinguistics. Whether large language models (LLMs) exhibit a functionally analogous distinction remains an open question, particularly given that LLMs rely on the same underlying mechanism (next-token prediction) for both input and output processing. In this exploratory study, we operationalize the production-perception distinction through direct token probability measurements rather than metalinguistic prompting. Using the base Llama-3.1-8B model, we generated poems under a production prompt and re-scored the same tokens under both rephrased production prompts and perception-oriented prompts. Across an extended experiment with four production and three perception prompts, production-perception distances consistently and substantially exceeded production-production distances, with non-overlapping ranges across conditions and an overall average ratio of approximately 1.8. Near-ceiling correlations in the production-production control confirm that the effect is specific to communicative framing rather than prompt surface variation, and we show the effect replicates across five open-weight models (Llama-3.1-8B, EuroLLM-9B, gemma-2-9b-it, Mistral-7B-Instruct-v0.3, and Qwen2.5-7B-Instruct), spanning both base and instruction-tuned variants. Temporal analysis revealed that the perception prompt exerts its strongest influence at the beginning of the sequence, with divergence decaying as generated context accumulates, though the specific shape of this decay varies across prompt pairs. These findings suggest that prompt framing alone induces a production-perception distinction in LLM probability distributions, even within a decoder-only architecture.
arXiv:2607.10986v1 Announce Type: new
Abstract: This paper introduces a hierarchical control architecture for multi-agent adversarial environments, decoupling strategic task planning from rigorous safety assurance. The system formulates pursuit-evasion as a zero-sum receding-horizon game, solved via an iterative minimax \acl{mpc} scheme. This allows pursuers to anticipate and block evader trajectories using transverse velocity penalties rather than relying on reactive heuristic formations. To guarantee collision-free operation without compromising the convexity of the \acl{mpc}, a discrete-time \acl{cbf} operates as an inner-loop safety filter. Through simulated experiments, we demonstrate the framework's adaptability. By simply altering the weights of the shared zero-sum payoff and \acl{cbf} constraints, the swarm can fluidly switch from aggressive pursuit-evasion tactics to strict perimeter defense and area denial, demonstrating robust performance across varying rules of engagement without structural changes to the control logic. The source code is available: https://github.com/ananya-ac/pursuit-evasion-mpc-cbf.
arXiv:2607.06723v2 Announce Type: replace-cross
Abstract: Adaptive optimizers carry hidden states that change how visible gradients become parameter motion. We develop optimization geometrodynamics as a variational theory of this hidden geometry. Infimal pushforward eliminates all hidden states realizing the same action and composes across optimizer hierarchies. Under smooth nondegeneracy, it yields hidden susceptibility and the Schur-complement curvature seen after relaxation. For affine pre-reduction perturbations, the induced interaction curvature is the negative-semidefinite operator $-G^*H^{-1}G$, whose mixed entries integrate to finite mechanism contrasts.
Our main realization is the determinant-one affine-invariant SPD action map $P\mapsto PA$. We prove a global analytic bundle with closed totally geodesic fibers and a unique analytic nearest-controller section. A strongly convex fiber theorem and an explicit logarithmic action residual give a globally linearly convergent solver from every feasible initializer, together with nonasymptotic value, controller-distance, and residual bounds and observable posterior stopping certificates. A conditional inexact result propagates supplied rigorous residual-error and radius majorants. The dense spectral kernel is confined to an active subspace of dimension $r\le 2m$, yielding an explicit spectral-arithmetic operation bound and a strict dimensional reduction when $r<d$. For nested shape-normalized quadratic actions, canonical multi-secant projections satisfy an exact CAT(0) Pythagorean decrease and recover the determinant-one inverse Hessian shape at the sharp rank threshold $d-1$, provided the scalar gauge $c_H=(\det H)^{1/d}$ is known. These results turn the action bundle into an exact iterative computation with posterior certificates and a finite-identification theory.
arXiv:2607.10028v1 Announce Type: new
Abstract: Turbulent bluff-body wakes exemplify the coexistence of large-scale coherent structures and fine-scale turbulence -- two ends of a wide dynamical range of scales connected through the turbulent cascade. In this work, we study the multiscale dynamics in the high-Reynolds-number wake behind a circular disk. One-point and two-point statistics are first examined, including the budget and spectra of the turbulent kinetic energy (TKE). Streamwise advection is found to contribute the most to the TKE balance, while the dissipation rate does not follow the classical equilibrium scaling ($\varepsilon \nsim \mathcal{U}^3/\mathcal{L}$). The largest scales are represented by the three-dimensional coherent modes extracted using spectral proper orthogonal decomposition, whereas the TKE and Reynolds shear stress spectra exhibit inertial-range scalings. A filtering-based triple decomposition further separates the fluctuations into large- and small-scale components and partitions the kinetic energy, with respective spatial transports at each scale and an inter-scale transfer in between. The inter-scale fluxes indicate a statistical forward cascade and follow the classical $\mathcal{U}^3/\mathcal{L}$ scaling, while their radial profiles become self-similar. The disequilibrium between inter-scale flux and dissipation is shown to arise from non-negligible streamwise advection at the sub-filter scale. Finally, the observed anti-correlation between the dissipation coefficient and the local Taylor Reynolds number, $C_\varepsilon = \varepsilon \mathcal{L}/\mathcal{U}^3 \sim Re_\lambda^{-1}$, is shown to originate from a similar correlation in the coarse-grained, locally averaged statistics. The results suggest that the instantaneous cascade-dissipation disequilibrium is intrinsic to turbulence and becomes apparent when large-scale unsteadiness and length-scale growth prevent statistical equilibrium.
arXiv:2607.10034v1 Announce Type: new
Abstract: Large language models (LLMs) store factual knowledge in their parameters. While recent work has shown that this knowledge resides in MLP layers, existing constructive and mechanistic interpretability models of fact-storage in LLMs fail to explain the surprising empirical phenomenon that they store facts at an information-theoretically optimal rate. In this work, we develop a theoretical account of this phenomenon. We develop the first Transformer-compatible fact-storing MLP closed-form construction that satisfies the following three properties empirically observed in LLMs: it (i) attains optimal fact storage scaling, (ii) handles arbitrary input/output geometries, and (iii) works inside Transformers. Key to our work is to analyze the decoding margin of MLPs, whereas prior work only studies MLP fact storage. Under isotropic embeddings, our construction achieves information-theoretically optimal storage capacity scaling and requires $10$-$104\times$ fewer parameters at matched fact count than prior constructions. For arbitrary key and value embeddings, we show that our construction attains the same storage capacity scaling, up to penalization factors depending on the embedding geometries. Moreover, we demonstrate that our constructed MLPs can be used within Transformer blocks for factual recall tasks at optimal capacity scaling, requiring $15$-$63\times$ fewer parameters at matched fact count than prior constructions. Finally, as a proof-of-concept, we show that fact-storing MLPs enable modular fact editing by swapping a Transformer's MLP with a new one.
arXiv:2607.11109v1 Announce Type: new
Abstract: Statute retrieval is a fundamental task in legal information retrieval, yet existing approaches struggle to bridge the gap between colloquial legal queries and formal statutory language. In this paper, we propose GCSR, a generative statute retrieval framework that reformulates statute retrieval as a sequence generation problem and internalizes statutory knowledge into a generative model. Specifically, we propose a multi-granularity structured docid that encodes legal hierarchy and semantic information, together with a multi-task training strategy. Experiments show that GCSR consistently outperforms strong sparse, dense, and legal-domain baselines. Our results demonstrate the effectiveness of generative retrieval for statute retrieval and highlight its potential for broader legal information access and downstream legal reasoning tasks.
MMA-Former: Multi-Window Mixture-of-Head Attention Transformer for Adaptive PNI Prediction in 3D MRI
arXiv:2607.10988v1 Announce Type: new
Abstract: Perineural invasion (PNI) is a critical prognostic factor in cholangiocarcinoma. Non-invasive prediction from 3D MRI is challenging, demanding models that efficiently capture both fine-grained details and global context. We propose the Multi-window Mixture-of-Head Attention Transformer (MMA-Former), a novel end-to-end 3D architecture featuring a Coarse-Fine Transformer (CFT) structure for parallel multi-scale feature extraction. We advance this structure by integrating a novel Window-Specific Mixture-of-Head attention (WS-MoH) mechanism. Unlike standard Multi-Head Self Attention (MSA), WS-MoH generates a representation for each 3D window and dynamically routes the entire window to specialized or common attention heads. This enables spatially adaptive feature extraction tailored to the local context of each window, enhancing specialization and reducing redundancy without increasing parameters. Evaluated on a retrospective dataset of 168 T1-weighted MRI scans, MMA-Former achieved an AUC of 0.752, outperforming other 3D architectures, including the best CNN (AUC of 0.708) and Transformer baselines (AUC of 0.681).
arXiv:2607.11127v1 Announce Type: new
Abstract: Organizations and regulators increasingly consult large language models (LLMs) for regulatory-compliance questions, yet a wrong statutory citation can silently propagate into legal advice, compliance documentation, and policy decisions. We introduce a bilingual benchmark of 120 questions probing whether freely accessible LLMs fabricate article citations for two data-protection instruments: the EU General Data Protection Regulation (GDPR) and the Saudi Personal Data Protection Law (PDPL). The benchmark pairs direct citation retrieval questions with false premise verification probes and deliberately unanswerable "trap" questions -- including questions about a repealed article and about deadlines that exist only in implementing regulations, not in the law itself. Every question is posed in both Arabic and English, and all scoring is fully automatic against a manually verified gold reference. Evaluating three freely accessible models (Gemini 2.5 Flash, GPT-OSS-120B, Nemotron-3-Super-120B), we find a dramatic jurisdiction gap: near-ceiling citation accuracy on the GDPR (94-100% on direct retrieval) against majority fabrication on the Saudi PDPL (60-77%), invariant to query language; the highest fabrication rates (67%) arise from statute-vs-regulations confusion, and 91% of fabricated citations are asserted with confidence >= 0.8. Fabrication tracks the jurisdiction of the law, not the language of the query, and model confidence provides no protection -- indicating that verbatim-verification safeguards, rather than model self confidence, must gate any institutional reliance on LLMs for compliance screening.
arXiv:2607.10309v1 Announce Type: new
Abstract: Reinforcement learning (RL) is commonly employed to enhance the performance of autonomous systems, including the Autonomous Internet of Things (AIoT). However, the trial-and-error nature of RL, when conducted in real-world environments, is costly and hazardous in some scenarios. Consequently, the majority of RL research is conducted in simulation. This reliance introduces challenges related to the Sim-to-Real transferability. Evaluating the Sim-to-Real algorithmic robustness and the Sim-to-Real gap is a critical prerequisite for research aimed at improving RL performance in the real world. Therefore, industries such as robotics have developed concurrent simulation and physical platforms to facilitate this research. However, a universal Sim-to-Real benchmark platform for AIoT does not currently exist. To address these concerns, we developed a real-world AIoT platform for studying RL in AIoT. On this platform, an agent deployed on an edge device plays video games on a separate host computer via a hardware-emulated keyboard, guided by vision input. This platform uses commercially available components costing less than USD 400, together with two computers. Because the system's objective is game score maximization, it inherently mitigates safety risks associated with real-world RL deployments. Experimental results show the simulation-trained agent suffers a 1160% performance degradation relative to the human-level performance after real-world deployment, indicating a significant Sim-to-Real gap. Direct real-world training using the deep Q-network (DQN) algorithm achieves approximately 38% of human-level performance after 10 million training steps, demonstrating the feasibility of RL under real-world conditions. These results suggest that the proposed Sim-to-Real benchmark platform provides a substantial foundation for qualitative and quantitative evaluations of RL in real-world AIoT systems.
arXiv:2607.11128v1 Announce Type: new
Abstract: Real-time driving risk assessment provides an essential basis for proactive safety by identifying and quantifying the danger of ongoing road interactions before adverse outcomes occur. However, due to the scarcity of collision data and frame-level risk labels, existing driving risk assessment methods often rely on surrogate objectives, which may imperfectly align with true collision risk and not faithfully reflect the relative danger of driving interaction. This paper proposes a comparison-based ordinal risk learning framework that learns collision-relevant risk scores from pairwise supervision in driving data, directly modeling relative risk ordering without requiring numerical frame-level risk labels. We derive pairwise comparisons from three sources of event-structured driving data for such ordinal risk learning: temporal progression within safety-critical sequences, event-level contrast between dangerous and normal interactions, and physics-based counterfactual perturbations. On this basis, instantiations with three risk-scoring function parameterizations are implemented, including directly learning risk scores from comparison data, and aligning existing single or multiple surrogate-based risk models. The proposed framework is evaluated on the 100-Car and SHRP2 naturalistic driving datasets using a proactive collision warning task. Results show that the proposed framework improves high-recall risk discrimination, warning precision, and warning lead time over representative surrogate-based baselines across both in-distribution and out-of-distribution evaluations. These results suggest that the proposed framework can contribute to proactive safety research by providing more reliable risk assessment for automated driving systems and safety-critical driving interactions.
arXiv:2607.11152v1 Announce Type: new
Abstract: In this paper, we present an integrated terrestrial and satellite maritime communication system, where a shorebased terrestrial base station (TBS) and a low Earth orbit (LEO) satellite cooperatively provide wide-area communication services to maritime users. We conduct performance analysis for the integrated satellite-terrestrial maritime communication system. Specifically, we analyze the transmission rate and coverage probability of near-shore and off-shore users respectively according to the maritime communication environment. Besides, in order to better understand the impact of some key parameters, we also make asymptotic analysis in some special cases. Further, we design an optimization algorithm to maximize the coverage probability of near-shore users by adjusting the transmission power of TBS and the LEO satellite, while ensuring the both off-shore and near-shore users can meet the minimum communication rate requirements. Finally, extensive numerical analysis results verify the accuracy of the theoretical results and the effectiveness of the proposed optimization algorithm in the maritime communication system.
arXiv:2607.10508v1 Announce Type: new
Abstract: LLM agents that conduct research (proposing ideas, writing and running code, analyzing results) can already carry a study from research question to figures, yet cannot be fully trusted. The same question asked twice in a row returns different answers; the agent announces a number that no execution produced, and tool use does not prevent this, because nothing binds what the agent reports to what its tools returned; a small upstream change leaves downstream results silently stale, with no way to list which ones; and the agent re-runs preprocessing and rewrites code it has already produced. We argue these failures share one root: every step of today's agent loop is a stochastic LLM call whose internal state nobody, including the agent, can check. Rather than trying to see inside the LLM, we take a lesson from databases, which earn trust without being watched, because deterministic operators over well-defined state make their guarantees hold by construction. We propose organizing a research project the same way. The project lives in a deterministic, versioned dataflow engine (in effect, a query plan over materialized views), and the LLM, together with the user, is a stochastic compiler that may only edit that plan. The executor never calls the LLM; LLM output enters only as versioned code and data that the executor then runs, and any asserted result enters the record only with an execution behind it. Five design rules at this boundary turn familiar database machinery, from versioning and provenance to incremental maintenance and cost-based scheduling, into guarantees that make research reliable, non-wasteful, transparent, and collaborative. This report presents the diagnosis, the requirements, and the design; the guarantee walkthrough, a prototype, and the research agenda appear in the full version, in preparation. The LLM, we argue, should be the query compiler, never the executor.
arXiv:2607.10567v1 Announce Type: new
Abstract: Concurrent game frames are a standard semantic framework for logics of strategic reasoning. Two notions of coalition power can be derived from such frames: alpha powers and actual powers. An alpha power of a coalition is a set of possible futures such that the coalition has an action that forces the resulting future to lie in that set. An actual power of a coalition is a set of possible futures satisfying the following condition: the coalition has an action such that (1) the action forces the resulting future to lie in the set, and (2) every future in the set is compatible with that action. Recent generalizations of concurrent game frames separate three structural assumptions built into the standard model: seriality, independence of agents, and determinism. This yields eight classes of general concurrent game frames. In this paper, we prove that for actual powers, the four classes of general concurrent game frames, where independence of agents is not assumed, are representable by four corresponding classes of neighborhood frames. Building on this result, we show that for alpha powers, the same four classes of general concurrent game frames are likewise representable by four corresponding classes of neighborhood frames.
arXiv:2607.11170v1 Announce Type: new
Abstract: Multimodal anomaly detection benefits from complementary RGB and 3D evidence, yet auxiliary RGB reconstruction is not equally reliable across product categories and class-wise test-time policy selection is usually unavailable. We propose TC-MAF, a base-anchored multi-evidence fusion design that combines a multimodal detector, complementary Dinomaly evidence, and a small cross-modal consistency cue under one fixed pixel-level fusion formula. A lightweight training-dispersion confidence (TDC) term scales auxiliary participation using only normal training statistics. On MVTec-3D, TC-MAF reaches 0.979 image-level AUROC and 0.990 pixel-level AUPRO, achieving the best mean results on both detection and localization among the compared multimodal methods. Systematic ablations show that the fusion structure itself is the dominant factor, while TDC provides a smaller but reproducible calibration gain over no calibration or arbitrary calibration. Additional experiments show that the same design remains effective under a pooled-statistics variant, auxiliary-branch and backbone substitutions, few-shot settings, a missing-3D setting, and cross-dataset evaluation on Eyecandies. Code is available at https://anonymous.4open.science/r/TC_MAF-C3BB.