arXiv:2607.10235v1 Announce Type: new
Abstract: Previous work in group recommender systems has demonstrated a sensitivity to the distribution of preferences within a group. Specifically, the selection of the preference aggregation strategy benefits from considering such group configurations. In this paper, we study whether LLMs are able to mimic this sensitivity and to select the ideal aggregation strategy (and corresponding recommendation) according to nuanced human perceptions of fairness, satisfaction, and consensus.
We do this by fine-tuning Large Language Models (LLMs) on human survey data to serve as real-time judgmental models within the recommendation pipeline. Using a reasoning dataset distilled from DeepSeek-V3.1 and human ground truth assessments, we develop Judgmental Llama and Judgmental OLMo to simulate group assessments. Our pipeline successfully generates multiple recommendation candidates based on social choice-based aggregation strategies and dynamically selects the one that maximizes these predicted human-like evaluations. We further validate these suggestions in a user study (n=284) and find that our methodology achieved the highest scores for satisfaction and group consensus. Furthermore, we find that LLM judgments are most aligned with human perceptions of fairness, satisfaction and consensus when we also consider interaction effects between our LLM-based method and group configuration (e.g., minority or coalition). These findings give further support for dynamically adapting aggregation strategies to specific within-group preference distributions, and highlight the advantage of using LLMs for an adaptation that is aligned with subjective human judgments.
Science Journals
arXiv:2607.10844v1 Announce Type: new
Abstract: Codes over non-unitary rings have been studied recently. In particular, codes over the commutative non-unitary ring $I_p$ (in the classification of Fine) of order $p^2$ where $p$ is a prime are being considered. For $p=2$ (resp. $p=3$), three categories of codes over $I_p$ have been studied: self-orthogonal codes, quasi self-dual codes, and self-dual codes over $I_p$. Using some related mass formulas and building-up constructions, classifications of these codes have been done up to the permutation equivalence (resp. the monomial equivalence) for certain small lengths. In this paper, we take the prime $p=5$ and consider the ring $I_5$. We introduce the notion of linear codes over $I_5$. We also define the same three categories of linear $I_5$-codes, study the structures of these $I_5$-codes and relate them to their associated residue and torsion codes. We classify the three categories of codes completely in lengths at most $4$ up to the monomial equivalence for a given type $\{ k_1 , k_2 \}$. Moreover, in the paper of Alahmadi et al. regarding the mass formula for self-orthogonal codes over $I_p$, mistakes in the classification of quasi self-dual codes over $I_5$ had been made such as incorrect automorphism group order of some codes or inconsistency with the mass formula for self-orthogonal codes over $I_p$ for length $n=2$ and type $\{ 1 , 0 \}$ and for length $n=3$ and type $\{ 1, 1 \}$. We correct and improve such results.
arXiv:2607.10768v1 Announce Type: new
Abstract: LLM-based agents are increasingly deployed to solve optimization problems, yet existing benchmarks evaluate them on pre-structured mathematical formulations that bypass the most critical challenge: translating complex business requirements into correct models and solve efficiently. We introduce Opti-Agent-Bench, an end-to-end benchmark that evaluates Large Language Models (LLMs) across the complete optimization R&D pipeline, from understanding business-language descriptions through mathematical modeling, algorithm selection, and code implementation, to solution report generation. Our design rests on three pillars: (1) businesssemantic authenticity with anti-template traps that defeat pattern matching; (2) modular evaluation with cross-module consistency checking across Problem Understanding, Formal Modeling, Implementation, and Reporting; and (3) the ORAC bi-level validity framework that simultaneously ensures task quality and scoring integrity. Across several industrialscale tasks spanning integer programming, robust optimization, stochastic programming, and non-convex optimization, we expose critical failure modes of current models, including constraint omission, model-code inconsistency, and report-implementation divergence, that remain invisible under conventional single-metric evaluation.
arXiv:2607.10788v1 Announce Type: new
Abstract: Pre-Charge Half Buffer (PCHB) is a promising asynchronous digital design paradigm for harsh-environment operation; however, its soft-error characteristics remain largely unexplored. This paper presents a systematic soft-error characterization and hardening trade-off analysis for static PCHB circuits. A controlled transistor-level fault-injection framework is developed to extract polarity-dependent critical charge at internal nodes. Vulnerability nodes are identified based on extensive simulation. Four mitigation strategies, double-sided Schmitt trigger, single-sided Schmitt trigger, transmission-gate reinforcement, and duplication-based redundancy, are implemented and evaluated across five representative PCHB cells. Comprehensive resilience-overhead comparisons in delay, energy, and area are reported, leading to architecture-specific hardening guidelines for robust static PCHB design.
arXiv:2607.10523v1 Announce Type: new
Abstract: Data narratives increasingly shape public understanding, but their failures are rarely just isolated factual errors or deceptive charts. Instead, they emerge through a broader meaning-making process in which quantitative evidence is transformed into claims, representations, and arguments. While prior work has examined these failures across disparate fields (e.g., statistics, visualization, and fact-checking), the community lacks a holistic lens to explain how these issues arise, propagate, and compound. To address this gap, we introduce TIC, a Taxonomy of Issues in Data Communication, synthesized from prior literature and refined through the qualitative annotation of 700 real-world data narratives from fact-checking sites, research datasets, and controversial media. TIC organizes recurring breakdowns across six dimensions-data, analysis, visual encoding, text, reasoning, and interpretation-and situates them within a framework spanning analysis, narrative construction, and audience reception. Alongside the taxonomy and process framework, we contribute a qualitatively annotated case corpus with coding justifications and an interactive browsing interface. Collectively, these contributions provide a structured lens for diagnosing problematic data narratives and informing future sociotechnical support for trustworthy data communication.
arXiv:2607.10649v1 Announce Type: new
Abstract: Coverage path planning (CPP) is a fundamental problem in robot motion planning, whose aim is to produce robot trajectories that provide complete coverage of target workspaces while minimizing task-specific objectives such as path length, overlap, number of turns, and energy consumption. CPP has widespread applications in cleaning, inspection, mapping, agriculture, manufacturing, surveillance, demining, and environmental monitoring. Although classical CPP has been extensively studied, recent advances have extended CPP beyond single-robot settings to multi-robot systems, complex 3D environments, constrained platforms, learning-based coverage planning, and visual coverage tasks. This paper presents a comprehensive survey of 125 representative works published primarily between 2015 and 2026, while presenting the evolution of recent developments in light of the classical CPP methods published before 2015. The CPP methods are organized into six main categories: single-robot CPP, multi-robot CPP, 3D CPP, constrained CPP, learning-based CPP, and visual CPP. For each category, the review summarizes the main planning formulations, representative algorithms, strengths, and limitations. In addition, the review analyzes how environmental knowledge, workspace geometry, robot constraints, sensing objectives, and coordination requirements shape the CPP problem. The survey further discusses open challenges in scalable online planning, multi-robot coordination, 3D and visual coverage, unified platform-constrained and resource-aware coverage, and learning-enhanced coverage. Thus, the survey provides a structured overview of recent CPP developments and future research directions.
arXiv:2607.10387v1 Announce Type: cross
Abstract: Temporal grounding in long recordings remains challenging for audio-conditioned LLMs. We present a time-aware audio LLM that answers questions with explicit timestamps over up to 120 minutes of input. Our approach interleaves periodic time markers with continuous audio tokens using large-scale synthetic supervision from a cascaded pipeline. Our model achieves strong temporal-grounding accuracy on short and long benchmarks and supports time-anchored fragment descriptions and summaries. Extensive ablations examine how time representation, marker frequency, tokenization, and duration-mixture design affect accuracy and computational cost. We release model weights and datasets to support further research on time-aware audio understanding, available at https://huggingface.co/ai-sage/GigaChat3.1-Audio-10B-A1.8B.
arXiv:2607.10572v1 Announce Type: new
Abstract: Mutual correlated agreement captures whether a random linear combination of received words can create a new large agreement with a code, a property relevant to the soundness of batched proximity testing. We show constructively that list-decoding counterexamples yield lower bounds on the mutual correlated agreement error. Given an explicit counterexample to the $(p,L)$-list-decodability of a linear code over $\mathbb{F}_q$, we construct a related code $C'$ of the same length and dimension such that $\operatorname{err}_{\mathrm{MCA}}(C',p)\ge\frac{1}{q}\left\lceil\frac{(L+1)q}{q+L}\right\rceil$, while decreasing its minimum distance by at most one. The construction also produces an explicit pair of words witnessing this error.
We further give a structure-preserving version for code families whose coordinates are indexed by a finite set $\Omega$, with each index determining a generator-matrix column through a map $v:\Omega\to\mathbb{F}_q^k$. The construction changes at most one coordinate index and ensures that the output code remains in the same indexed family. As applications, we instantiate this principle for algebraic-geometry (AG) evaluation codes and Reed--Solomon codes. For AG codes, if $G$ is the divisor defining the underlying Riemann--Roch space and $N$ is the number of rational places outside $\operatorname{supp}(G)$ available for evaluation, the resulting code remains over the same function field and Riemann--Roch space, with a modified set of evaluation places. Its mutual correlated agreement error is at least $\frac{1}{q}\left\lceil\frac{(L+1)N}{N+L\mathrm{deg} G}\right\rceil$. The Reed--Solomon conclusion follows as the Vandermonde-column specialization.
arXiv:2607.10805v1 Announce Type: new
Abstract: On-Policy Self-Distillation (OPSD) has emerged as a crucial paradigm for enhancing and aligning Large Language Models (LLMs). However, in complex reasoning tasks, OPSD paradoxically degrades downstream performance. In this paper, we systematically investigate this pathology and identify a severe optimization trap we define as \textbf{Thinking Collapse} -- a sharp decline in the model's native intermediate reasoning behavior, measured by epistemic-token density (ET per 1k). Through entropy-based gradient masking and token-level target analysis, we show that this collapse is triggered by aggressive teacher gradients at high-student-entropy decision forks, where student epistemic tokens are frequently suppressed into teacher non-epistemic targets and are highly concentrated in high pointwise student-teacher divergence regions. To resolve this optimization pathology, we propose \textbf{Adaptive Dual-Perspective OPSD (AD-OPSD)}, a robust control framework that dynamically moderates the self-distillation objective. AD-OPSD selectively anchors high-suppression-risk sandboxed tokens to a reference prior derived from the frozen base model via an asymmetrical pointwise divergence gate, preserving native thinking capacity while retaining OPSD's error-correcting power. Extensive experiments across competitive mathematical benchmarks show that AD-OPSD improves over standard OPSD by up to \textbf{+4.1\%} absolute average accuracy across diverse model scales and datasets. Further analysis demonstrates that AD-OPSD mitigates thinking collapse and generalizes robustly to different post-training paradigms.
arXiv:2607.10369v1 Announce Type: new
Abstract: Flow-matching policies have emerged as an effective policy parameterization for robot learning. They iteratively generate actions from noise, enabling highly expressive modeling of complex and multimodal action distributions. However, prior works observed that scaling these policies with value-gradient reinforcement learning (RL) often leads to training instability. Existing methods attribute this instability to iterative generation and therefore avoid end-to-end value-gradient optimization by sacrificing iterative generation, high expressiveness, or value-gradient optimization. Contrary to prior belief, we show the instability does not stem from iterative generation itself, but from the vanilla sampling strategy originally designed for behavior cloning, which becomes brittle under value-gradient RL. Motivated by this insight, we propose VINE, an RL-oriented sampling method that enables stable end-to-end value-gradient optimization for flow-matching policies. Instead of following a single flow trajectory, VINE reconstructs a new interpolation state at every denoising step, creating a stable differentiable path for value-gradient propagation while remaining compatible with the original flow-matching denoising process. As a result, VINE preserves the expressiveness and iterative generation of flow-matching without sacrificing end-to-end value-gradient optimization. Despite performing end-to-end backpropagation through all ten denoising steps, VINE achieves stable policy improvement and consistently outperforms state-of-the-art RL methods on the OGBench offline RL benchmark and real-world robotic manipulation task. Videos are available on our website: https://agibottech.github.io/vine.
arXiv:2607.10591v1 Announce Type: new
Abstract: Non-binary bottom-up constituency parsing is usually taken to require arity actions: reductions such as \(\textsc{Reduce-}X\#k\) specify both the mother label and the number of children to be composed. We show that this arity parameter is not a necessary transition primitive. Our parser introduces constituent labels separately and recovers reduction spans from delimiter-bounded stack configurations. In a well-formed reduction configuration, arity is uniquely determined by the active delimiter and the label marker, making it a derived property of parser state rather than an action label. This factorization removes label--arity-specific reduce actions while preserving direct construction of original non-binary trees. Experiments on PTB and CTB show that the delimiter-guided parser remains competitive with an arity-specific bottom-up baseline under the same implementation framework, with substantially smaller action inventories. Analyses further show that its predicted arity profile remains close to the gold treebanks and that high-arity constituents do not collapse when arity actions are removed.
arXiv:2607.10031v1 Announce Type: new
Abstract: This paper presents a maximally flat-top half-band FIR optical filter on a silicon nitride (SiN) platform at 193 THz. Using a cascaded Mach-Zehnder Interferometer (MZI) topology simulated in Lumerical INTERCONNECT, MODE, and FDTD, the filter achieves insertion losses of 0.1349 dB and 0.1761 dB with extinction ratios of 18.317 dB and 23.002 dB for Channel-A and Channel-B, respectively, under realistic S-parameter conditions.
arXiv:2607.10798v1 Announce Type: new
Abstract: Multimodal retrieval-augmented generation (RAG) is often evaluated with clean evidence, yet real retrieval can return topically relevant but unreliable content: false text and misleading images from corrupted metadata, entity swaps, typographic overlays, semantic edits, adversarial patches, blends, or style transfer. We introduce QIMG-7, a controlled benchmark for multimodal retrieval pollution in multi-sentence factual QA, spanning four datasets, seven image-attack families, and 16 paired clean/polluted regimes, for 1,760 evaluation rows per method. Across four generator/gate stacks, naive multimodal fusion is brittle: in the main gpt-4o-mini stack, Full-MM support drops from 0.908 with clean text to 0.490 with polluted text, often making Parametric fallback safer than retrieval. We propose source-aware trust resolution (SATR), a training-free approach that compares Parametric, Text-only, and Full-MM candidate answers and selects among candidate answers or falls back based on source reliability. The Field-Selector variant achieves the best balanced score, 0.816, improving over Full-MM by 11.7 points and over the Cascaded Router by 2.7 points. Ablations show that, in this text-first setting, explicit text-reliability modeling is the dominant driver of these gains. Overall, in text-first factual QA with multimodal retrieval conflict, our results support selective trust rather than unconditional fusion. Artifacts are available at https://github.com/SaadElDine/Trust_Before_Fusion.
arXiv:2607.09869v1 Announce Type: new
Abstract: The Shapley value is a widely used concept in attribution problems, as it uniquely satisfies the axioms of linearity, consistency, equal treatment, and efficiency. Often, the inclusion AUC metric is used to evaluate the quality of player rankings, in order to identify positively participating players. However, it can be established that the Shapley value is not always reliable for this purpose. The core issue lies in its linearity: the Shapley value acts as a linear operator with an excessively large null space, which is likely to contain non-negligible perturbations that remain indistinguishable to the operator. To address this limitation, we explore the design of nonlinear axiomatic attribution methods. Inspired by the least core, which is a popular nonlinear substitute for the Shapley value, we introduce a class of nonlinear attribution methods that retain the remaining necessary axioms. Each method yields a contribution vector that is the unique optimal solution to a minimization problem, which aims to approximate utility functions as faithfully as possible. In terms of the inclusion AUC metric, our experiments demonstrate the potential effectiveness of these methods compared to Shapley value variants that relax only the efficiency axiom. Our code is available at https://github.com/watml/nonlinear-axiom.
arXiv:2607.10926v1 Announce Type: new
Abstract: Identifying heterogeneous treatment effects under unobserved confounding is central in observational causal inference. In proxy models with a discrete latent confounder, prior Synthetic Potential Outcomes (SPO) [Mazaheri-Squires-Uhler '25] recover the mixture of treatment effects through recursively constructed scalar moments. We show that this sequence is one projection of a more fundamental object. Under the same population factorization assumptions, there is an exact compressed observable operator: after projecting onto the shared proxy signal subspace, the difference of two treatment-arm quotient operators is similar to the diagonal matrix of latent treatment effects. Its eigenvalues are the latent effects; its lifted left eigenvectors, after anchor normalization, recover the target-proxy feature matrix and then the latent mixture proportions. Every scalar SPO moment is a bilinear functional of a power of this operator. The resulting estimator handles overcomplete proxy systems, replaces high-order scalar inversion with finite-dimensional spectral analysis, and admits high-probability first-order perturbation bounds for treatment effects, feature rows, and simplex-projected mixture weights.
arXiv:2607.10033v1 Announce Type: new
Abstract: Identifying the causes of Earth's extremes is challenging because counterfactual experiments are not possible in the observed world, while numerical experiments are computationally expensive and subject to biases. Data-driven causal discovery offers a complementary path, but existing approaches can fail in undersampled, high-dimensional regimes, and may not recover multi-timestep, multivariate pathways leading to particular events. We introduce Tracer of Causal Evolutions in Space and Time (TraCE-ST), a probabilistic Lagrangian approach that produces event-conditioned causal trajectories in multivariate gridded data. In synthetic experiments and real-world extreme events, TraCE-ST recovers known causal drivers and estimates their relative contributions, while also highlighting less-studied drivers, including orography-driven vorticity for Tropical Storm Debby (2006) and anomalous ocean-surface fluxes for the 2021 Pacific Northwest heatwave. Here, we propose causal tracking as an efficient data-driven framework for synthesizing causal evidence and generating testable hypotheses, complementing association analyses and numerical modeling while accelerating the study of high-impact events.
arXiv:2607.10074v1 Announce Type: new
Abstract: Graph machine learning provides powerful tools for understanding complex networks and learning meaningful node representations. A central challenge, however, is designing embeddings with minimal distortion of both local and global functionals, such as shortest path lengths. Prior distortion guarantees for distance-preserving embeddings are worst-case in nature, producing overly pessimistic bounds that fail to capture the structure of typical large-scale networks. To address this, we analyze shortest-path approximation via landmark-based embeddings on inhomogeneous random graphs, a general model with type-dependent edge probabilities. By retaining shortest paths to a small set of reference nodes called landmarks, landmark-based methods effectively function as virtual graph spanners, where structural heterogeneity and controlled neighborhood expansion modeled via multi-type branching processes enable significantly tighter dimension-distortion trade-offs than classical worst-case bounds. We extend these guarantees to global, component-wide averages and unify the analysis across finite-type and continuous latent spaces through a novel metric sandwiching framework, establishing universal distortion bounds for general $L^2$ kernel models, including heavy-tailed and power-law networks. Finally, we introduce a GNN-augmented variant that replaces rigid, computationally expensive exact shortest-path queries with flexible, structure-aware neural surrogates. By leveraging the inherent alignment between graph neural message-passing and the dynamic programming principles of shortest-path algorithms, our approach demonstrates that models trained on small-scale random graphs learn to extract universal distance-preserving features, achieving robust generalization to large-scale, real-world networks that match or exceed the fidelity of classical, exact landmark-based embeddings.
arXiv:2607.10080v1 Announce Type: new
Abstract: Large language models (LLMs) are increasingly being used as automated judges
for relevance evaluation in information retrieval, yet their robustness to
adversarial manipulation remains insufficiently understood, particularly in
multilingual settings. In this work, we investigate the impact of
cross-lingual prompt injection attacks on LLM-based relevance judgments using
TREC Deep Learning collections and two open-weight models under established
prompting frameworks. We examine both instruction-based and content-based
injection strategies in 8 languages spanning different resource levels. Our
results demonstrate that multilingual query-based injections are highly
effective in inflating relevance scores while simultaneously evading existing
prompt-injection defenses. We further found that, although existing defense
mechanisms can be modified to mitigate such attacks, these injections can be
easily adapted to bypass them. These findings highlight a critical gap in
current defense approaches and demonstrate that language generalization can
act as an attack vector, underscoring the need for more robust and proactive
evaluation frameworks for LLM-as-a-judge systems.
arXiv:2607.10831v1 Announce Type: new
Abstract: We present a least-squares weak Galerkin (LS-WG) finite element method for solving the Stokes equations on arbitrary polygonal and polyhedral meshes. By utilizing discrete weak derivatives on discontinuous polynomial spaces, the proposed framework naturally accommodates complex domain geometries and general partitions. Crucially, this least-squares formulation bypasses the traditional inf-sup (LBB) compatibility condition, transforming the standard indefinite saddle-point problem into an inherently symmetric and positive definite (SPD) discrete linear system. We establish the well-posedness of the numerical scheme and rigorously derive optimal-order error estimates in a custom discrete energy norm. Specifically, we prove convergence rates of $\mathcal{O}(h^k)$ for the discrete projection error and $\mathcal{O}(h^{k-1})$ for the global approximation error when employing polynomials of degree $k \ge 1$ for the velocity field and $k-1$ for the pressure. Extensive numerical experiments confirm these theoretical convergence rates, demonstrating the method's robustness, geometric flexibility, and overall efficiency.
arXiv:2607.10070v1 Announce Type: cross
Abstract: We study maximum-sum red-blue matchings and matching equilibrium for finite planar point sets. For a red-blue perfect matching $M = \{(a_i,b_i) : 1 \le i \le n\}$, we define the gain of a directed red cycle as the change in total weight produced by cyclically shifting the corresponding blue partners. We prove that $M$ is maximum-sum if and only if every directed red cycle has nonpositive gain, and we derive a geometric sufficient condition for optimality from cyclic intersections of distance-difference regions. We then characterize balanced matchings, in which all red-blue perfect matchings have the same total weight. Equilibrium is shown to be equivalent to vanishing cycle gains, to an additive form of the distance matrix, and to a common level-set condition for distance-difference functions. In the squared Euclidean case this yields an orthogonality classification, while in the Euclidean case it yields a hyperbolic level-set description and a collinear-separation classification in the nondegenerate setting.
arXiv:2607.10882v1 Announce Type: new
Abstract: Mode-selective PT-symmetry is the manifestation of Vernier-assisted cross-injection between oscillators with unequal delays. PT-symmetry has been proposed as a mechanism for achieving low-noise single-mode operation in optoelectronic oscillators, but existing formulations are typically based on matched delay loops and frequency-independent coupling, leading to symmetry transitions that are global in frequency and therefore not intrinsically mode selective. A more general formulation of PT-symmetric time-delay oscillators is developed in which the coupling operator is allowed to be dispersive. This permits frequency-selective PT-symmetry transitions and removes the requirement for matched delay loops. Two mathematically equivalent but physically distinct realisations are derived. The first corresponds to coupled gain-loss loops connected by a dispersive coupler, while the second corresponds to a pair of equal-gain oscillators coupled through symmetric cross-injection with unequal delays. Numerical simulations confirm the predicted behaviour and demonstrate strong sidemode suppression while preserving the low phase-noise characteristics of large-delay oscillators. The results establish a direct connection between PT-symmetry, cross-injection architectures, and Vernier oscillator design.
arXiv:2607.10185v1 Announce Type: new
Abstract: This study aims to verify whether combining non-thermal equilibrium atmospheric pressure plasma with benzoyl peroxide (BPO) can enhance the inhibition of bacteria. This work involves not only adjusting the parameters of the plasma device, such as increasing the power supply voltage, modifying the working gas flow rate, as well as incorporating a specific proportion of oxygen to the working gas, but also aims to combine plasma with BPO to inactivate bacteria. An optical emission spectroscopy was utilized to determine the optimal working gas flow rate and oxygen mixing ratio for the experimental setup. Additionally, surface temperature and ultraviolet (UV) emission measurements were also performed to characterize the physical properties of the plasma treatment. Although the precise chemical mechanisms between plasma and BPO have yet to be fully elucidated, bacterial inactivation experiments on Escherichia coli demonstrated that the addition of BPO indeed enhanced the sterilization efficacy of plasma treatment.
arXiv:2607.10187v1 Announce Type: new
Abstract: Large language models (LLMs) write Unity C\# for game scenes. Yet nearly all demonstrations rest on an iterative repair loop that regenerates code until it compiles, conflating what the model writes with what the loop fixes. We remove the loop and evaluate a single pass, where the first draft is final. This isolates the model's parametric knowledge, the most stringent test of unaided generation. Models instantiate Goal Playable Concepts, playable counterparts of goal patterns, across 10,400 generations (four open-weight models, 7B--30B; two generation modes; four intermediate-representation (IR) conditioning levels; 26 goal patterns; 20 seeds). None compiled into a runnable scene, leaving no survivorship bias. To understand how the generated C\# scripts fail, we categorize the 99 error codes behind 90{,}673 compiler-error occurrences as Grounding (invented or misused Unity types and APIs) or Hygiene (structural defects needing no Unity knowledge). The split differs sharply by goal pattern (e.g., Stealth fails mostly on invented engine references; Capture on plain C\# structure). Larger models, stricter IRs, and different generation modes move the errors but never yield a compiling scene. The bottleneck is missing engine-specific knowledge. The census orders goal patterns by that demand, showing designers where single-pass generation breaks.
arXiv:2607.10610v1 Announce Type: new
Abstract: Efficient waste segregation is critical for sustainable urban management and environmental governance. Existing automated systems are limited by single-modality visual processing, insufficient contextual understanding, and weak regulatory alignment. To address these issues, we propose a language-guided vision-AI framework that integrates vision-language models and multimodal large language models for joint visual-linguistic reasoning. This framework implements a visual question answering paradigm aligned with India's Solid Waste Management Rules 2016. We construct a new WasteVQA dataset with 13,500 question-answer pairs across 21 waste categories. Experiments show that the BLIP-based model achieves a BLEU score of 0.8291 and a BERTScore of 0.9273, outperforming traditional CNN-based methods. This work improves source-level segregation accuracy, ensures regulatory compliance, and supports scalable deployment for municipal and citizen-facing waste management, promoting multimodal AI in sustainable urban infrastructure. The source code and dataset are available at: https://github.com/Khushkataruka/WasteAssistant
arXiv:2607.10403v1 Announce Type: cross
Abstract: We propose a novel muon Moonshot concept for lunar subsurface tomography based on upward-going muons originated from the lunar regolith. Unlike the Earth, the Moon lacks an atmosphere, leaving a dense regolith below and a near-vacuum environment above. Consequently, while most downward-going hadrons are absorbed before decaying, upward-going hadrons escaping the regolith can decay in flight, producing a significant source of lunar muons. These muons are detectable by instruments on the lunar surface or in near-lunar orbit. We perform Monte Carlo simulations to investigate their energy spectra, angular distributions, and integrated fluxes under various theoretical and detector configurations. The results indicate that the lunar muon flux is sensitive to detector altitude under a flat-terrain assumption, demonstrating its potential as a novel non-invasive probe of shallow subsurface voids. We also present case studies on detecting underground cavities and water resources, with cavity-induced flux variations observable in less than a minute and weaker water signals distinguishable after about ten minutes of data collection, and discuss potential implementations in future lunar missions.