arXiv:2607.15086v1 Announce Type: new
Abstract: We study time-discrete parametric approximations of evolution equations in Hilbert spaces based on residual minimization. The solution is represented by a parametrized ansatz belonging to a low-dimensional nonlinear manifold, and time stepping is performed by minimizing suitably defined residuals at each step. Two natural residual formulations are considered: discretization followed by parametrization of the evolution equation, and discretization of the Dirac--Frenkel variational principle governing the parameter dynamics. A unified error analysis is developed for both approaches within the family of $\zeta$-methods. The resulting bounds separate the effects of time discretization from those of residual minimization and yield first- and second-order convergence under Lipschitz, one-sided Lipschitz, and dissipativity assumptions. For the variational formulation, additional stability conditions involving the conditioning of the parametrization map arise naturally. The framework is applied to Gaussian approximation manifolds, for which residual norms and gradients admit explicit closed-form expressions when polynomial operators are involved. This enables efficient implementation without spatial discretization. Numerical experiments for time-dependent Schr\"odinger equations illustrate the theoretical convergence rates and the influence of residual accuracy on conservation properties.
Science Journals
arXiv:2607.14497v1 Announce Type: new
Abstract: Egocentric Visual Question Answering (VQA) has attracted widespread attention as an important task for enabling Multimodal Large Language Models (MLLMs) to interact with the real world. However, existing MLLMs struggle to perform effective spatial reasoning in complex egocentric scenes due to their limited spatial perception capabilities. To this end, we introduce Ego Scene Augmentation (ESA), an egocentric spatial perception framework, which actively enhances the spatial perception capabilities from the egocentric perspective, powered by the proposed Ego-element Graph. Our core insight is leveraging the Ego-element Graph as an intermediary representation to augment the egocentric spatial perception of MLLMs via visual foundational models. Specifically, we 1) construct the Ego-element Graph, which encapsulates and integrates egocentric spatial features enabled by visual foundational models; 2) enhance the spatial perception capabilities of MLLMs via the Ego-element Graph for ego-perspective scenes. Our proposed ESA framework presents significant performance improvement on the EgoTextVQA benchmark. We achieve an 8.14% gain on the indoor setting and an 8.72% gain on the outdoor setting. Furthermore, our ESA shows the most impressive performance improvement in the shopping subset of the indoor setting. The project code is publicly available.
arXiv:2607.15109v1 Announce Type: new
Abstract: Optical frequency combs are key components of several photonics applications including spectroscopy, communications, and ultrafast photonics. A central challenge in frequency-comb photonics is to develop sources whose operating state can be precisely controlled and adapted to different application needs. We introduce frequency comb functionality to a THz metasurface vertical-external-cavity-surface-emitting laser (VECSEL), combining its characteristic high output power and excellent beam quality with a reconfigurable comb output. The source exhibits reversible switching between actively mode-locked 3.5 ps-long pulses and stable frequency-modulated quantum walk comb states. The flexible control of the intermodal phase relation is achieved through careful dispersion engineering via a Gires-Tournois interferometer (GTI) output-coupler combined with resonant RF bias modulation of the metasurface. These results pave the way for on-demand comb control in the THz range and provide a versatile strategy that could be extended to other semiconductor frequency-comb platforms and wavelength ranges.
arXiv:2607.15111v1 Announce Type: new
Abstract: Vehicle coordination at unsignalized intersections relies on accurate real-time vehicle state acquisition and reliable command-and-control (C&C) signal delivery. However, existing studies typically treat sensing, communication, and control separately, which may lead to redundant transmissions, outdated state information, and unreliable vehicle coordination. In this paper, we investigate a new scenario of distributed integrated sensing and communication (ISAC)-enabled vehicle coordination at intersections, where multiple roadside units (RSUs) collaboratively transmit sensing signals for vehicle state acquisition and C&C signals for vehicle movement control under the management of a central base station (BS). To improve signaling efficiency, we propose a unified goal-oriented semantic communication (GSC) framework, which transmits sensing and C&C signals only when they are semantically important for improving intersection traffic throughput. Specifically, an extended Kalman filter (EKF) is adopted to predict vehicle states and fuse distributed sensing measurements. A masked hybrid proximal policy optimization (MHPPO) framework is then developed to jointly determine sensing transmission decisions, C&C transmission decisions, and C&C signal contents based on a value-of-information (VoI) reward. Furthermore, we propose an uncertainty-aware transmission design (UTD), including robust beamforming and VoI-based time-division power allocation, to improve sensing and communication reliability under vehicle state uncertainty and inter-RSU interference. Simulation results show that our proposed framework achieves 100% collision-free vehicle coordination with significantly reduced signaling overhead compared with predictive ISAC baselines adapted from state-of-the-art related studies and several ablation baselines.
arXiv:2607.14499v1 Announce Type: new
Abstract: Multi-modal Large Language Models (MLLMs) have made substantial advances on benchmarks, yet their real-world effectiveness remains uncertain. This gap stems from the fundamental misalignment between benchmarks in controlled, static settings and the dynamic, interactive, and contextualized nature of real-world applications. To bridge this gap, we propose CEDI (Contextualized Evaluations of MLLMs through Dynamic, multi-round Interactions), a framework that recasts evaluation as a three-party interaction between an evaluatee model, an automated examiner, and a grader. The examiner conducts multi-turn, semi-structured conversation guided by a graph-based representation of the task. By navigating state-space transitions, CEDI deploys diverse strategies, from clarification requests to adversarial probes, to elicit performance evidence. We apply CEDI to visual hallucinations. Empirical results across multiple models, diverse settings, datasets, and domains show that contextualized, interactive evaluations reveal not only significantly more hallucinations than conventional static evaluation but also ones that more closely resemble those arising in practical use cases. We further show that hallucinations often accumulate over long contexts, through self-reinforcing dialogue history, and models are particularly vulnerable to questions requiring premise rejection or refusal. Together, these findings highlight CEDI as a step toward realistic, systematic, and ecologically valid assessments of MLLMs' capabilities. Code is available at github.com/williamium3000/cedi.
arXiv:2607.15114v1 Announce Type: new
Abstract: Recommender systems increasingly shape which content reaches users, making it important to understand whether coordinated activity is amplified beyond the accounts that initiate it. Existing robustness evaluations largely focus on static target-rank changes and do not capture how coordinated interactions, recommendation, and user response evolve within a feedback loop. To address this gap, we propose CoSimRec, an offline agent-based evaluation framework that models coordinated accounts, dynamic ranking, non-bot responses, and ranking interventions in a shared closed-loop process. CoSimRec introduces the Algorithmic Penetration Rate (APR) metric family to measure target content's share of non-bot exposure and engagement, lift against matched no-attack baselines, and exposure gained per coordinated interaction. We evaluate CoSimRec on MIND, MovieLens, and LastFM using random, popularity-based, feedback-sensitive, MF, and BPR-MF recommenders, with ten-seed inference for the primary APR analysis and population-scale experiments of up to 1000 users. Random controls show no statistically supported positive penetration, whereas popularity-based and feedback-sensitive ranking produce significant positive APR-Lift in all six master-worker dataset--recommender settings, reaching 0.4505 on LastFM; synchronization-aware ranking reduces APR in every corresponding defense setting.
arXiv:2607.14233v1 Announce Type: new
Abstract: Physics-informed neural networks (PINNs) have had a broad research impact in modeling domains governed by partial differential equations (PDE). However, PINNs have been shown to perform poorly, sometimes even converging to trivial solutions, in challenging PDE domains, or when generalizing to unseen but related PDE domains. Previously proposed solutions detail hyperparameter tuning to reduce loss imbalance between data-driven and physics guided losses, curriculum learning based training strategies, or dynamic re-sampling of hard collocation points. These methods face certain pitfalls: hyperparameter tuning is expensive, designing a training curriculum is ambiguous in multi-parameter PDE settings, and dynamic resampling still fails in complex PDE settings. Complementary to this line of thinking, we believe the initial PINN network weights also play a crucial role in the emergence of catastrophic failures during training, yet the effect of PINN weight initialization has been surprisingly under-investigated. To this end, we propose a framework for Learned Initialization via Gated Layerwise Optimization (LIGO-PINN) to overcome PINN convergence failures. Through rigorous evaluation on 1D and 2D PDE domains, including a challenging 2D fluid dynamics setting, we demonstrate that our methodology outperforms state-of-the-art methods designed to alleviate PINN failures, achieving a 91.5% average performance improvement across six baselines and 81% over the strongest baseline. We also verify that LIGO-PINN generalizes to 3D unstructured domains. Finally, we analyze training dynamics across all three PDE domains to explain both LIGO-PINN's improvement and the convergence failure of traditional PINNs. Code: https://github.com/scailab/ligo-pinn
Keywords: Machine Learning, Physics-Informed Neural Networks, Deep Learning, PDE Modeling
arXiv:2607.15118v1 Announce Type: new
Abstract: Side-channel attacks pose a significant security threat for modern computing platforms, because they exploit subtle discrepancies in CPU behaviors to leak sensitive information. To model the information leaked by a CPU via microarchitectural side-channels, recent work proposed leakage contracts: an ISA-level security abstraction that provides the foundations for secure CPU programming. Unfortunately, due to the complexity of current microarchitectures, devising a leakage contract for a CPU requires extensive manual effort and thus modern CPUs lack dedicated leakage contracts.
We present a methodology to extract instruction-centric leakage contracts for major CPU architectures with minimal manual intervention. We implemented this technique in malcos, the first template-free tool that automates the synthesis of leakage contracts for black-box CPUs. We evaluate malcos on x86 and ARM CPUs, and show that the contracts it synthesizes are precise and sound with respect to all leaks observed during synthesis. Our results demonstrate that learning leakage contracts from black-box CPUs is feasible.
arXiv:2607.15123v1 Announce Type: new
Abstract: Recent FPGAs have improved deep learning (DL) inference efficiency through dedicated tensor blocks and in-BRAM computation. ReRAM-based analog in-memory computing (IMC) pushes efficiency further, offering an order-of-magnitude improvement in compute density and energy efficiency over conventional digital logic by performing vector-matrix multiplication (VMM) directly within the ReRAM crossbar; prior work has integrated such IMC blocks into FPGAs for DL inference. However, conventional IMC designs support only static-weight VMM, leaving nonlinear operations and dynamic matrix-matrix multiplication (DIMM) to the FPGA fabric. As a result, the benefits of IMC are largely confined to static-weight models, whereas Transformer-based models, which rely on frequent nonlinear and DIMM operations, gain only limited improvement. Moreover, the ADCs within each IMC block consume more than 70% of its area and power, further limiting system efficiency and scalability. To address these limitations, we propose a novel FPGA architecture that integrates an ADC-free IMC block, replacing the conventional ADC with analog content-addressable memories (ACAMs) that natively perform nonlinear operations inside the block. To fully exploit this block, we conduct an FPGA-aware design-space exploration that determines optimal crossbar dimensions while balancing FPGA area, flexibility, and DL performance, and we develop an efficient mapping that leverages ACAMs to carry out DIMM operations, extending the applicability of IMC to attention computation. On CNN and Transformer-based benchmarks, the proposed architecture achieves up to 40x and 1.9x higher energy efficiency and 4.1x and 2.5x higher area efficiency, respectively. Overall, it significantly improves FPGA DL inference efficiency and sustains robust gains on Transformer-based workloads across long input sequences, advancing domain-specialized FPGA design.
arXiv:2607.15128v1 Announce Type: new
Abstract: Hyperspectral image (HSI) classification requires reliable pixel-relation modeling under spectral variability, mixed pixels, and heterogeneous boundaries. Existing graph-based HSI classifiers usually construct graph topology from spatial proximity, superpixel connectivity, or learned feature affinity. However, the spectral physical prior carried by contiguous bands has limited influence on topology estimation and message propagation. This paper presents DAPGNet, a dynamic adaptive physics-guided graph diffusion network that injects a structure-constrained physical prior into relation-level graph learning. DAPGNet first encodes contiguous spectral responses into node-wise multiscale physical-prior representations. A two-stage graph constructor then combines spectral-spatial affinity, physical-prior consistency, and spatial distance to form a physical-prior-aware sparse topology. During graph diffusion, learned edge weights are transformed into additive attention biases, while a physical gate performs node-wise and feature-wise interpolation between graph-aggregated features and projected physical-prior features. Cross-scale fusion integrates node states from different diffusion depths, and the network is optimized with main classification, auxiliary supervision, and second-order spectral smoothness regularization. Experiments on Indian Pines, WHU-Hi-LongKou, Houston2013, and Houston2018 show that DAPGNet achieves the best OA, AA, and Kappa among representative CNN-, Transformer-, Mamba-, and graph-based baselines. It improves AA over the strongest competing method by 3.64 to 7.31 percentage points across the four datasets. Ablation and sensitivity analyses further support the complementary effects of physical-prior extraction, prior-aware topology construction, physics-gated propagation, and spectral smoothness regularization.
arXiv:2607.14095v1 Announce Type: new
Abstract: Retrieval Augmented Generation (RAG) has proven to be a widely successful process at improving the quality of outputs from a Large Language Model (LLM) for wider context. However, RAG systems typically retrieve context from flat document stores, which struggles when queries require hierarchical or relational reasoning across structured knowledge. I present HG-RAG (Hierarchy-Guided RAG), a framework that performs graph-traversal over a hierarchical knowledge graph to deliver structured context to a language model. My retrieval pipeline resolves a named entity anchor from the query, then expands context upward through parent nodes, laterally through relational neighbors, and downward through child nodes when needed. I evaluate HG-RAG against a dense retrieval baseline across three world scales (18-800 nodes) with four query types: local fact, hierarchical, neighborhood, and multi-hop. Results show HG-RAG consistently outperforms the flat baseline on hierarchical, relational, and multi-hop reasoning tasks, while reducing hallucination and maintaining locality coherence.
arXiv:2607.15134v1 Announce Type: new
Abstract: We study how a representative sample of United States adult AI-assistant users (n=1,999; June 2026) choose among platforms, allocate tasks across them, evaluate provider trustworthiness, and value data-handling features. Estimates are weighted to the AI-user population using external adoption benchmarks. Four patterns emerge. The market is concentrated but internally differentiated: ChatGPT is the primary assistant for 58% of users and Gemini for 25%, yet smaller platforms hold defensible task niches--Claude captures a third of coding tasks despite a 7% overall share. Task allocation is thus organized by platform far more than by user, and technical use falls steeply with age. Trust is earned through use rather than reputation: Claude is ranked most trustworthy in every head-to-head among users of both platforms, and shows by far the largest gap between how its users and non-users rate it. Finally, privacy concern is near-universal but action is gated by knowledge, not concern; in a choice experiment users pay most to keep humans--not models--out of their conversations ($11.20/month), with valuations rising in task sensitivity.
arXiv:2607.15141v1 Announce Type: new
Abstract: To numerically integrate a function, one may resort to Quasi-Monte Carlo estimators, that average integrand values at pseudo-random well-distributed uniform sampling locations. Better uniformity improves the worst-case integration-error bound. A standard measure of uniformity is given by an integer $t$ value, where $t=0$ yields the best uniformity. Producing sequences of samples with bounded $t$ values can be achieved with Sobol' recursive construction, that uses coefficients of irreducible polynomials. While $b$-dimensional sequences with $t=0$ can be obtained by taking $b$ polynomials of degree $1$ over the Galois Field $\mathrm{GF}(b)$, we show conditions that guarantee $t=0$ for specific higher degree polynomials. In particular, we relate the Sobol' construction to tensorized powers of Pascal matrices when the chosen polynomials only differ by a constant and exhibit simple conditions to guarantee $t=0$ in this case. We then focus on Artin-Schreier irreducible polynomials, in the form $p_i(x) = x^b - x + c_i$, where $i \in \{1, \dots, b-1\}$ and $b$ is prime, and we make explicit conditions that always guarantees $t=0$ in $b-1$ dimensions. Combining $b$-dimensional Sobol' of degree $1$ and our $(b-1)$-dimensional Artin-Schreier sequence of degree $b$, we provide a fast greedy procedure that optimizes the $(2b-1)$-dimensional combined $t$ value, while guaranteeing $t=0$ projection in subspaces.
arXiv:2607.14500v1 Announce Type: new
Abstract: Deep learning is a promising approach to optimize wireless communication by simplifying the search for near-optimal solutions. Prior studies on deep learning-based wireless communication optimization have explored supervised learning approaches that map raw user information, such as location or channel state information, to optimal power allocation vectors. While this approach demonstrates competitive performance, it is susceptible to adversarial attacks via input perturbations. Current defense mechanisms primarily rely on empirical methods, which do not provide formal guarantees of robustness. We fill this gap by proposing a formal verification framework to evaluate the robustness of deep learning-based power allocation in multi-cell massive multiple-input multiple-output (MIMO) systems against a wide range of potential adversarial input manipulations. To the best of our knowledge, this is the first attempt to formally verify deep neural networks in a regression setting with non-linear output constraints. We model the adversary's capabilities using hyper-rectangle constraints on their perturbation, adopt the abstraction-based bound-propagation technique (DeepPoly) to bound the interval of potential allocated powers, and formulate the minimum performance requirements as a constrained program for numerical feasibility analysis. Evaluation on publicly available datasets for power allocation in multi-cell massive MIMO indicates that a well-trained model can guarantee the local robustness under location perturbation by +-1m while retaining a maximum 1% optimality gap.
arXiv:2607.14096v1 Announce Type: new
Abstract: In predictive modeling, the ability to explain why a model produces a given target prediction has become increasingly important [5, 10]. Black-box models do not provide a transparent description of the internal mechanisms that generate the prediction, making even accurate predictions difficult to interpret and validate. In critical contexts, predictive accuracy alone is not a sufficient validation metric if the reasons underlying model decisions remain unexplained. The IMEX (Interaction-Based Model Explanation) approach represents a methodological direction within explainable predictive modeling. IMEX is designed to identify which variables contribute most to the target prediction and which interactions among variables are significant in determining the target. The method does not impose limitations on higher-order interaction analysis, allowing the investigation of feature subsets with cardinality greater than two. Beyond the identification of feature importance, IMEX enables the exploration of interaction patterns that may be consistent with latent mechanisms influencing the outcome. Through the application of the IMEX algorithm, it is possible to construct an interpretability map of the predictions. The IMEX framework is built on two complementary metrics: Static Correlation Power (PCS), which quantifies the contribution of individual features, and Interaction Correlation Power (PCI), which captures non-additive effects among features. In the present work, the PCS component is experimentally validated through a comparison with INVASE [18] on three synthetic datasets with known structures. The results indicate that IMEX can recover relevant feature-level structures in the presence of non-linear, conditional, and multicollinear relationships between input features and prediction targets.
arXiv:2607.14595v1 Announce Type: new
Abstract: Large-scale video diffusion models (VDMs) deliver strong generation performance, but full fine-tuning for downstream tasks incurs prohibitive computational costs. Existing parameter-efficient fine-tuning (PEFT) methods have two critical flaws on billion-scale models: they still require substantial trainable parameters, and reward-based training suffers from noise-induced optimization instability in condition-guided tasks. We propose MagicPrompt, a lightweight framework that achieves extreme parameter efficiency and stable reward optimization. It first adopts Attention-Embedded Prompt Tuning, which steers generation via lightweight soft prompts with orders of magnitude fewer parameters while preserving pre-trained knowledge. It further introduces Dual-Space Reward Feedback Optimization, which uses self-supervised latent objectives to improve condition-guided reward training. Experiments show MagicPrompt reaches competitive performance with less than 1\% trainable parameters and notably reduces training costs.
arXiv:2607.14416v1 Announce Type: new
Abstract: The interconnected nature of global financial systems makes them vulnerable to systemic risks, where the failure of a few institutions can trigger catastrophic cascading defaults. Traditional risk models often fail to capture the complex, non-linear dynamics of these networks. While Graph Neural Networks (GNNs) have shown promise in modeling relational data, they primarily learn correlative patterns and function as black boxes, offering little insight into the causal mechanisms of shock propagation. This limitation is critical for regulators who require explainable models to perform stress tests and devise effective interventions. We introduce CausalGraphX, a novel framework that integrates GNNs with counterfactual reasoning to provide explainable assessments of systemic risk. CausalGraphX employs a Graph Attention mechanism to learn representations of institutional vulnerability and uses an adversarial regularization technique to ensure these representations capture causal drivers rather than spurious correlations. Furthermore, we propose an optimization-based approach to generate counterfactual explanations, answering questions such as, "What minimum capital injection would have prevented Bank A's default under a specific stress scenario?" We validate CausalGraphX on large-scale synthetic financial networks. Our results demonstrate that CausalGraphX significantly outperforms traditional and deep learning baselines in predicting cascading defaults while providing sparse, plausible, and actionable counterfactual explanations.
arXiv:2607.14097v1 Announce Type: new
Abstract: We introduce RegNetAgents, an AI-oriented multi-agent framework for structured, query-driven regulatory candidate identification across heterogeneous gene regulatory networks. The system enables unified analysis of bulk tumor and single-cell-derived ARACNe networks by integrating TCGA-derived cancer networks with large-scale single-cell regulatory networks from the GREmLN project. For a given focal gene, the framework performs dual-network classification, cancer gene filtering using OncoKB annotations, and mode-of-action (MoA) assignment for tumor-derived regulatory relationships. Candidates are ranked by evidence consistency across networks (Both, TCGA-only, GREmLN-only). The system is implemented as a multi-agent LangGraph DAG workflow, accessible through a unified Python API and Model Context Protocol (MCP) client, operating as a downstream analytical layer over precomputed regulatory networks rather than a network inference method. Across eleven breast cancer (BRCA) and twelve colorectal cancer (COAD) focal genes, RegNetAgents identifies candidate regulators significantly enriched for OncoKB-annotated cancer genes. TCGA-derived candidates show strong enrichment (Stouffer Z = 6.69 for BRCA and 6.95 for COAD), while GREmLN-derived candidates also demonstrate significant enrichment (Z = 5.51 for BRCA and 7.06 for COAD; all p < 0.0001). No enrichment is observed in housekeeping or non-driver control gene sets, supporting signal specificity. An extended module enables structured evaluation of oncogenic potential, druggability, clinical relevance, and network vulnerability, supporting end-to-end interpretation from candidate identification to biological hypothesis generation. RegNetAgents establishes an interpretable AI framework for cross-network regulatory candidate identification in cancer genomics.
arXiv:2607.14346v1 Announce Type: new
Abstract: Policy learning methods are increasingly used to inform treatment allocation under budget constraints. Most proposed methods assume complete treatment data, yet applications frequently suffer from missingness that can bias estimates and lead to suboptimal policies. We address this gap by extending efficient estimators for average treatment effect (ATE) estimation to policy value and conditional average treatment effect (CATE) estimation under missing at random (MAR) and missing completely conditionally at random (MCCAR) treatment data. Through asymptotic efficiency analysis, we prove that the MAR estimator, which leverages partially-observed units, is both valid and more efficient than the MCCAR estimator when MCCAR assumptions hold. This result provides formal justification for preferring MAR-based estimation in policy learning under both missing data settings. Our comprehensive experiments using synthetic and semi-synthetic datasets confirm that correctly specifying the missingness mechanism is crucial: misspecified estimators remain biased regardless of sample size, while our estimators achieve near-oracle performance when assumptions are satisfied. Our work provides practitioners with theoretically grounded, empirically validated tools for robust policy learning in the presence of missing treatment data.
arXiv:2607.15142v1 Announce Type: new
Abstract: World models are usually evaluated as components of model-based reinforcement learning (MBRL) systems, while the world models themselves are rarely studied in isolation.
We examine five representative visual world-model agents in Atari Pong: DreamerV3, DIAMOND, TWISTER, Simulus, and STORM. After reproducing their training pipelines and matching the reported agent performance, we freeze the learned world models and evaluate them with a closed-loop rollout diagnostic: a policy trained separately from the corresponding MBRL agent interacts with each frozen model, and the generated video trajectories are inspected for visual and dynamical errors. Across all five models, the rollouts contain clear failures, including ball disappearance, incorrect ball motion, and invalid ball-paddle interactions.
Beyond visual trajectories, we further evaluate them with pixel-space zero-shot MBRL, where a new policy is trained entirely inside a frozen world model and then evaluated in the real environment. Across all five models, the resulting policies substantially underperform those produced by the corresponding original MBRL training pipelines. The gap is particularly large for DreamerV3, whose mean return drops from -5.5 to -20.9, near the minimum Pong return of -21.
We hypothesize that insufficient modeling of task-critical concepts, such as the ball in Pong, may contribute to these failures. We therefore propose Concept-Guided Spatial Regularization (CGSReg), an auxiliary pixel reconstruction loss applied to segmented concept regions. Experiments show that CGSReg improves both closed-loop rollouts and pixel-space zero-shot MBRL in DreamerV3, DIAMOND, and TWISTER. Its effects vary across the remaining models and evaluation metrics, indicating that CGSReg alone does not address all world-model bottlenecks.
arXiv:2607.14099v1 Announce Type: new
Abstract: Deploying Vision-Language Models (VLMs) in real-world settings requires not only strong visual reasoning but also stability under sustained conversational pressure. We introduce Just Keep Prompting (JKP), a multi-turn evaluation framework that measures VLM epistemic stability when users repeatedly challenge, question, or contradict a model's answer. JKP probes models for up to 10 follow-up turns using three strategies: Adversarial Negation (repeated rejection), Pure Socratic Interrogation (repeated calls to reassess certainty), and Context-Aware Socratic Summarization (reflecting the model's prior rationale back before asking for reconsideration). We evaluate GPT-4o, Gemini 2.5 Pro, and Qwen3-VL-30B on a subset of the STAR benchmark across 720 multi-turn runs. Aggregate accuracy changes modestly from Turn 0 to Turn 10, but trajectory-level analysis reveals substantial instability: correct answers regress, wrong answers recover, and many runs exhibit repeated answer flipping. Repeated prompting has bounded upside and often acts as a destabilizer rather than a reasoning aid. The effect is strongly model-dependent: Qwen3-VL-30B achieves the highest final accuracy but becomes confidently wrong under direct contradiction; Gemini 2.5 Pro is comparatively stable but token-expensive; GPT-4o is the most brittle and oscillatory. These findings reveal that multi-turn VLM evaluation captures not just additional reasoning but pressure-response profiles: how models trade off visual grounding, calibration, and conversational compliance under repeated challenge.
arXiv:2607.15143v1 Announce Type: new
Abstract: AI coding agents set up projects by reading documentation and installing the dependencies it lists, without verifying their names, sources, or known vulnerabilities. By editing only a README, requirements file, or Makefile, an attacker can redirect the agent to an untrusted registry, a known-vulnerable version, or a wrong-but-plausible name: documentation becomes a vector for code execution. We present the first systematic evaluation of package-install-time supply-chain attacks delivered through ordinary project-setup documentation across production coding-agent harnesses, probing frontier models on twelve scenarios in five attack classes, grounded in documented incidents. The same model catches an attack through one harness and installs it through another: install-time security rests on the harness-model combination, not the model alone. Agents catch blatant typosquats reliably, but plausible separator-confusion names (azurecore for azure-core) slip through, and how often depends on the harness-model pairing. Source-based attacks like registry redirection are missed almost everywhere. The source blind spot recurs on npm and Cargo, where nearly every model installs the untrusted dependency; name detection carries over less consistently across ecosystems. Security-oriented prompts recover part of the gap but only for the dimension they name; a deterministic pre-install check that verifies names, sources, and versions before any code runs closes most of it.
arXiv:2607.15146v1 Announce Type: new
Abstract: Online encyclopedias shape political opinion and, through it, democratic discourse. In late 2025, Grokipedia was released, an encyclopedia written entirely by the LLM Grok. One motivation behind the project was to provide an unbiased alternative to Wikipedia, which has faced accusations of "left-wing" and "liberal" bias. But does an encyclopedia written by an LLM deliver greater neutrality, or does it simply embed a different ideology? We conduct a large-scale political bias study on Grokipedia and Wikipedia, analysing 1,394 article pairs describing members of government for neutrality along nine expert-coded ideology dimensions employing four LLM judges, Grok, Claude, Mistral, and DeepSeek. As the LLMs could themselves be biased, we also investigate patterns in their judgments. We find all LLM-judges, including Grok, to rate Grokipedia less neutral than Wikipedia. Both encyclopedias are rated as portraying politicians favourably overall, but towards different ideological groups. Grokipedia particularly favours economically right-wing politicians and penalises socially liberal ones, while Wikipedia is rated as favourably biased towards the latter.
arXiv:2607.15155v1 Announce Type: new
Abstract: The dynamics of premixed flames propagating in two-dimensional and cylindrical channels are investigated using direct numerical simulations of the fully compressible reactive Navier-Stokes equations coupled with conductive heat transfer within the channel walls. The simulations employ a high-order numerical method, detailed chemical kinetics and transport models for a stoichiometric hydrogen-air combustion. The influence of heat losses during the early stages of flame propagation is examined for channels of different aspect ratios, with particular focus on tulip flame formation and its subsequent transition to distorted tulip structures. Heat losses are modelled considering convective heat transfer from the hot combustion products to the inner wall surface, thermal conduction heat transfer through the wall, and convective and radiative heat losses from the outer wall surface to the surroundings. The obtained results are compared with corresponding simulations under adiabatic wall boundary conditions. The simulations reproduce the principal features of flame dynamics observed experimentally, highlighting the combined influence of wall heat losses and geometric confinement on flame dynamics in confined channels.
arXiv:2607.14100v1 Announce Type: new
Abstract: We present the first application of pregroup grammar-based quantum compositional natural language processing (QNLP) to Arabic; a morphologically rich, free-word-order language whose structural complexity provides a uniquely demanding testbed for theories of meaning composition in quantum circuits. Our system converts Arabic sentences into quantum circuits whose topology mirrors grammatical structure: subjects, verbs, and objects become quantum gates, and the typed dependencies between them (the pregroup grammar) determine how those gates are wired together. We conduct three controlled experiments spanning word order, morphological tense, and verb sense disambiguation, comparing quantum circuit methods against classical baselines including AraVec (Arabic word embeddings) and AraBERT (a pre-trained Arabic transformer).