arXiv:2607.01211v2 Announce Type: replace
Abstract: Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches. Their leaderboard scores are increasingly used as evidence of coding-agent progress, but those scores can conflate runtime instability, benchmark-specific scoring rules, and how many tasks are already solved by at least one public submission. We audit these issues across the three benchmarks. First, we replay the official reference patches for 740 code optimization tasks across four common types of Google Cloud machines. Most benchmark tasks can be replayed, but their reference patches satisfy the original benchmark validity rules in every cross-machine replay for only 39/102 GSO tasks, 11/140 SWE-Perf tasks, and 411/498 SWE-fficiency tasks; SWE-Perf is especially fragile because many reference patches produce close-to-zero runtime changes. Second, we show that public submission rankings depend strongly on the benchmark scoring rule. Among eight public submissions shared by GSO and SWE-fficiency, the official rankings disagree on 9 of 28 pairwise submission comparisons, and SWE-fficiency's leaderboard scoring rule assigns the worst ten tasks overly high score weights of 58.5%-82.8%. Third, looking across 10 public submissions for each task, we find that at least one submission matches or beats the reference patch on 85.3% (384/450) of replay-valid GSO and SWE-fficiency tasks, and beats the unoptimized base code on 99.8% (449/450). Our study complements leaderboard scores by identifying tasks with more reliable performance signals, quantifying per-task score contributions, and exposing the remaining performance gaps that are hidden by aggregate rankings.
Science Journals
arXiv:2607.15254v1 Announce Type: new
Abstract: Traffic agencies now have access to large volumes of video-derived data for studying safety and congestion. Most of these data are observational and collected without interventions, which makes causal questions such as "How would rain change traffic density?" difficult to answer. We present teLLMe, a system for exploratory causal analysis of urban driving datasets. The system starts from a structured event table built from dashcam annotations and combines causal structure learning with the PC algorithm, bootstrap-based stability checks, and query-specific effect estimation using linear regression and DoWhy. Natural-language questions are mapped to structured causal queries through a schema-aware LLM, enabling users to specify treatments, outcomes, and subpopulations. teLLMe returns a "Causal Card" that summarizes effect estimates, adjustment sets, DAG support, and assumptions, followed by a short natural-language explanation. Case studies on BDD-derived traffic events show that the system can surface plausible relationships involving weather, peak hours, and traffic density, while making uncertainty and modeling choices explicit. The system is designed as a tool for hypothesis generation and expert reasoning rather than a source of definitive causal claims.
arXiv:2607.14799v1 Announce Type: cross
Abstract: We derive a closed-form analytical expression for the distribution of eccentricities (DoE) in random regular graphs (RRGs) that consist of $N$ nodes of degree $c$. The DoE is given by the tail distribution $P(E > \ell) \simeq 1 - \exp \left[ - \exp \left( - \frac{ e^{b \ell} - \mu }{\beta} \right) \right]$, where the distance $\ell$ takes integer values, $b = \ln (c-1)$ is the shape parameter, $\beta = \frac{c-2}{c} N$ is the scale parameter and $\mu = \frac{c-2}{c} N \ln N$ is the location parameter. By providing the full distribution rather than a single characteristic length scale, we present a detailed view of the large-scale structure. In spite of the fact that the degrees of all the nodes are the same, their eccentricities exhibit non-trivial variations. We derive a closed-form expression for the mean eccentricity, which is given by $\langle E \rangle \simeq \frac{\ln N}{\ln (c-1)} +
\frac{\ln \ln N}{\ln (c-1)} - \frac{ \ln c - \ln (c-2) }{ \ln (c-1) } + \frac{1}{2}$. We calculate the mode of the DoE, which exhibits a staircase profile as a function of the network size. Interestingly, the mode is given by $E_{\rm mode} ={\rm Round} \left( \langle E \rangle \right)$, where ${\rm Round}( x )$ is the nearest integer to $x$. We also calculate the variance ${\rm Var}(E)$ and show that it exhibits oscillations as a function of the network size $N$. The results presented in this paper may serve as benchmarks for algorithmic approaches to eccentricity calculations in large sparse networks. The eccentricities are important in practical applications such as broadcasting and global dissemination, where the network performance is determined by the longest delay times.
arXiv:2607.15188v1 Announce Type: new
Abstract: We calculate radiative-recoil contribution of order $Z^2\alpha(Z\alpha)^5(m/M)^2m$ to the Lamb shift in muonium. This correction is due to insertion of radiative photons in the heavy line in the two-photon exchange diagrams. Our calculations are inspired by a new round of precise $1S-2S$ and $2S-2P$ experiments currently in progress.
arXiv:2607.15277v1 Announce Type: new
Abstract: In-context learning is commonly interpreted as a form of conditional inference, in which the prompt specifies a context and the model's output is treated as an estimate of the corresponding conditional distribution. If this interpretation holds, then LLM estimates should satisfy basic probabilistic identities. In particular, the law of total probability asserts that prior-weighted conditional distributions aggregate into population-level marginals over any valid partition of the population. In this work, we investigate to what extent LLM estimates adhere to this self-consistency principle. We use binary trees as an evaluation scaffold to recursively partition a population into increasingly fine-grained subpopulations. We then prompt LLMs with verbalized subpopulation descriptions in context, aggregate the resulting estimates back into population-level estimates, and compare them across partitions of varying granularity. Applying this protocol across problem domains and state-of-the-art frontier models, we show widespread violations of basic consistency properties. An in-depth study of persona prompting reveals a pattern we call the macro fallacy: estimates reconstructed from more fine-grained subpopulation responses are often better aligned with human reference data than direct population-level estimates. This effect persists across variations in tree structure and estimation task, and can be partially recovered through implicit prompting. Together, these findings suggest that models possess relevant subpopulation knowledge but do not reliably propagate it into aggregate estimates. This gap establishes statistical self-consistency as an unsaturated, reference-free criterion for evaluating LLMs.
arXiv:2607.14801v1 Announce Type: cross
Abstract: Hands-on telescope experience is often used to drive student engagement in astronomy education, but scaling access to larger groups of students is operationally challenging. Consequently, students encounter only a fraction of the professional workflow, rarely engaging with the rigorous peer-review, time-allocation processes, or automated data reduction pipelines that govern modern research facilities. We present the design of MIRA (Mentored Investigations using Robotic Astronomy), a data management and educational platform that connects Swiss secondary school and undergraduate students with operational robotic observatories. MIRA structures the entire observation lifecycle: proposal, review, acceptance/rejection, scheduling, and observation. Following execution, the platform automatically reduces raw FITS frames (including astrometric calibration and photometry) and serves them via a web-accessible archive accompanied by Python-based analysis tutorials. By separating educational front-ends from low-level telescope controls through Astra and ASCOM Alpaca, MIRA delivers an authentic scientific research workflow that bridges classroom learning with professional observatory operations.
arXiv:2403.10360v2 Announce Type: replace
Abstract: Lagrangian methods continue to stand at the forefront of the analysis of time-dependent dynamical systems. Most Lagrangian methods have criteria that must be fulfilled by trajectories as they are followed throughout a given finite flow duration. This key strength of Lagrangian methods can also be a limitation in more complex evolving environments. It places a high importance on selecting a time window that produces useful results, and these results may vary significantly with changes in the flow duration. We show how to overcome this drawback in the tracking of coherent flow features. Finite-time coherent sets (FTCS) are material objects that strongly resist mixing in complicated nonlinear flows. Like other materially coherent objects, by definition they must retain their coherence properties throughout the specified flow duration. Recent work [Froyland and Koltai, CPAM, 2023] introduced the notion of semi-material FTCS, whereby a balance is struck between the material nature and the coherence properties of FTCS. This balance provides the flexibility for FTCS to come and go, merge and separate, or undergo other changes as the governing unsteady flow experiences dramatic shifts. The purpose of this work is to illustrate the utility of the inflated dynamic Laplacian introduced in [Froyland and Koltai, CPAM, 2023] in a range of dynamical systems that are challenging to analyse by standard Lagrangian means, and to provide an efficient meshfree numerical approach for the discretisation of the inflated dynamic Laplacian.
arXiv:2405.00549v4 Announce Type: replace
Abstract: A Confirmation Rule is an algorithm run by network nodes to determine whether a block will remain permanently in the canonical chain. The only Confirmation Rule currently available in Ethereum's consensus protocol, Gasper, is FFG finalization. While it tolerates asynchronous network conditions, it is slow: in the best case, a transaction takes 13 to 19 minutes to confirm, depending on when it is submitted.
We devise a Fast Confirmation Rule (FCR) for Gasper that, under synchrony and the assumptions stated in this paper, achieves a best-case confirmation time of 12 seconds, a single slot, providing an order-of-magnitude improvement over FFG finalization. The rule is complementary to finalization: users who trust synchrony obtain fast confirmations, while finalization remains available as a fallback that tolerates asynchrony. Gasper is an ebb-and-flow protocol: it combines LMD-GHOST, a fork-choice rule providing fast progress under synchrony, with FFG-Casper, a finality gadget providing finality under partial synchrony. The main technical difficulty is to reason jointly about these two components, so that a block confirmed by LMD-GHOST cannot be filtered out by FFG-Casper's rules. We prove that the rule satisfies both safety, confirmed blocks remain canonical, and monotonicity, a confirmed block remains confirmed at all future times.
arXiv:2606.08322v2 Announce Type: replace
Abstract: This methodological study analyzes the effects of collinearity, effective dimensionality, and cluster stability in a 2023 study of US airline profit cycles from 1995 to 2020 by Renold et al., which uses k-means clustering, principal component analysis, and system dynamic modelling.We replicate their clustering experiment in three spaces -- the original 7-dim. raw-variable space, a 3-dim. PC score space, and a 4-dim. PC score space using their dataset. We show that the six-cluster taxonomy is geometrically robust: k-means in 3-PC space produces bit-for-bit identical cluster assignments relative to 7D raw space. As a nonlinearity check we apply kernel PCA under six kernels spanning three families plus a linear baseline. The kernels confirm an intrinsically linear manifold with no detectable curvature. The silhouette criterion reveals that the dataset structurally supports only three clusters, not six. Collinearity in the raw 7D space suppresses the silhouette signal. A kernel ridge regression check confirms no nonlinear accuracy gain over linear ridge once the COVID19 year is excluded. Together, these results argue for clustering on PC scores rather than raw variables in collinearity-prone panel data.
arXiv:2406.10593v3 Announce Type: replace
Abstract: Fine-tuning large language models (LLMs) for specific domain tasks has achieved great success in Text-to-SQL tasks. However, these fine-tuned models often face challenges with multi-turn Text-to-SQL tasks caused by ambiguous or unanswerable questions. It is desired to enhance LLMs to handle multiple types of questions in multi-turn Text-to-SQL tasks. To address this, we propose a novel data augmentation method, called QDA-SQL, which generates multiple types of multi-turn Q\&A pairs using LLMs. In QDA-SQL, we introduce a method incorporating validation and correction mechanisms to handle complex multi-turn Text-to-SQL tasks. Experimental results demonstrate that QDA-SQL enables fine-tuned models to exhibit higher performance on SQL statement accuracy and enhances their ability to handle complex, unanswerable questions in multi-turn Text-to-SQL tasks. The generation script and test set are released at https://github.com/mcxiaoxiao/QDA-SQL
arXiv:2606.09331v2 Announce Type: replace
Abstract: Omni-modal retrieval promises a single embedding space for text, image, video, document, and audio inputs, but building such a unified retriever is difficult since these modalities differ in data distribution, architecture, and optimization dynamics. In this work, we present Conan-embedding-v3, a decouple--fuse--recover framework for omni-modal retrieval. Conan-embedding-v3 first trains modality specialists independently and fuses their task vectors into a single dense backbone, a strategy we call Decoupled Specialist Fusion. We show that this fusion composes visual, video, and document retrieval capabilities, but also exposes a failure mode for projector-based modalities: when audio is attached through an external encoder and projector, fusing the backbone leaves the projector calibrated to the audio-specialist backbone, causing a large audio retrieval regression despite copying all audio-specific modules unchanged. We call this failure Projector Drift. To repair it, Conan-embedding-v3 applies Projector Recovery (i.e., full-parameter fine-tuning of the projector while keeping the backbone frozen) followed by balanced multi-modal rehearsal. The resulting model supports these retrieval pathways in one backbone, achieving 74.9 scores on MMEB while obtaining 55.61 on the 30-task MAEB audio suite.
arXiv:2603.13800v2 Announce Type: replace
Abstract: Visual spatial intelligence is critical for medical image interpretation, yet remains largely unexplored in Multimodal Large Language Models (MLLMs) for 3D imaging. This gap persists due to a systemic lack of datasets featuring structured 3D spatial annotations beyond basic labels. In this study, we introduce an agentic pipeline that autonomously synthesizes spatial visual question-answering (VQA) data by orchestrating computational tools such as volume estimation and bounding boxes extraction with multi-agent collaboration and expert radiologist validation. We present SpatialMed, the first comprehensive benchmark for evaluating 3D spatial intelligence in medical MLLMs, comprising 31,253 question-answer pairs across multiple organs and tumor types. Our evaluations on 24 state-of-the-art MLLMs and extensive analyses reveal that current models lack robust spatial reasoning capabilities for medical imaging.
arXiv:2603.15177v3 Announce Type: replace
Abstract: We introduce and investigate the irreducibility of endomorphisms of finitely generated free semigroups, i.e., we investigate when an endomorphism $\varphi: \Sigma^+ \to \Sigma^+$, where $\Sigma$ is any alphabet, can be nontrivially expressed as a composition $\varphi = \psi_2 \circ \psi_1$ of endomorphisms $\psi_1, \psi_2: \Sigma^+ \to \Sigma^+$. We, hence, study a notion of primality in the endomorphism monoid of the free semigroup -- a natural and fundamental concept in this algebraic structure. We establish that irreducibility is a nontrivial property for the class of so-called rank-preserving endomorphisms, and we provide a characteristic condition separating the reducible and irreducible endomorphisms. We also characterise when an endomorphism is a factor of another endomorphism, analyse the non-uniqueness of factorisations of a rank-preserving endomorphism into its irreducible components, and investigate the use of incidence matrices to give insights into the (ir-)reducibility of rank-preserving endomorphisms.
arXiv:2603.16351v2 Announce Type: replace
Abstract: Accurate taxonomic identification of parasitoid wasps within the superfamily Ichneumonoidea is essential for biodiversity assessment, ecological monitoring, and biological control programs. However, morphological similarity, small body size, and fine-grained interspecific variation make manual identification labor-intensive and expertise-dependent. This study proposes a deep learning-based framework for the automated identification of Ichneumonoidea wasps using a YOLO-based architecture integrated with High-Resolution Class Activation Mapping (HiResCAM) to enhance interpretability. The proposed system simultaneously identifies wasp families from high-resolution images. The dataset comprises 3556 high-resolution images of Hymenoptera specimens. The taxonomic distribution is primarily concentrated among the families Ichneumonidae (n = 786), Braconidae (n = 648), Apidae (n = 466), and Vespidae (n = 460). Extensive experiments were conducted using a curated dataset, with model performance evaluated through precision, recall, F1 score, and accuracy. The results demonstrate high accuracy of over 96 % and robust generalization across morphological variations. HiResCAM visualizations confirm that the model focuses on taxonomically relevant anatomical regions, such as wing venation, antennae segmentation, and metasomal structures, thereby validating the biological plausibility of the learned features. The integration of explainable AI techniques improves transparency and trustworthiness, making the system suitable for entomological research to accelerate biodiversity characterization in an under-described parasitoid superfamily.
arXiv:2607.14947v1 Announce Type: cross
Abstract: Modern generative models are increasingly trained using model-generated signals, creating both opportunities for self-improvement and risks of collapse. We study optimal self-distillation (SD) for rectified flow (RF): given a suboptimal teacher velocity field, can a student trained on a mixture of true RF velocities and teacher velocities provably improve the teacher? For linear RF with ridge regularization on fixed interpolation pairs, we prove an exact affine path identity, derive the optimal mixing coefficient in closed form, and show strict improvement in integrated velocity risk whenever the teacher risk is nonstationary along the regularization path. The optimal coefficient obeys a sign rule: positive mixing corrects under-regularized teachers, while negative mixing corrects over-regularized teachers. We also give one-shot generalized cross-validation (GCV) and validation tuning procedure that avoids grid search over mixing weights and repeated refitting. Combining this theorem with RF Wasserstein convergence bounds, we show that optimal self-distillation improves the velocity estimation terms controlling continuous-time and finite-step generation error. Experiments with Gaussian models, Gaussian mixtures, and image data show that optimal self-distillation improves velocity risk, mode recovery, and finite-step generation relative to both the teacher and pure distillation.
arXiv:2501.13703v2 Announce Type: replace
Abstract: Transfer Learning (TL) is an emerging field in modeling building thermal dynamics. This method reduces the data required for a data-driven model of a target building by leveraging knowledge from a source building. Consequently, it enables the creation of data-efficient models that can be used for advanced control and fault detection & diagnosis. A major limitation of the TL approach is its inconsistent performance across different sources. Although accurate source-building selection for a target is crucial, it remains a persistent challenge.
We present GenTL, a general transfer learning model for single-family houses in Central Europe. GenTL can be efficiently fine-tuned to a large variety of target buildings. It is pretrained on a Long Short-Term Memory (LSTM) network with data from 450 different buildings. The general transfer learning model eliminates the need for source-building selection by serving as a universal source for fine-tuning. Comparative analysis with conventional single-source to single-target TL demonstrates the efficacy and reliability of the general pretraining approach. Testing GenTL on 144 target buildings for fine-tuning reveals an average prediction error (RMSE) reduction of 42.1 % compared to fine-tuning single-source models.
arXiv:2508.00923v3 Announce Type: replace
Abstract: Large language models (LLMs) are increasingly used to answer health-related questions and support healthcare workflows, yet evidence for their safety still relies heavily on static benchmarks that can rapidly become obsolete or be optimized against. Here we introduce a Dynamic, Automatic, and Systematic (DAS) red-teaming audit framework that continuously stress-tests LLMs for health across four safety-critical axes: robustness, privacy, bias/fairness, and hallucination/factual inaccuracies. Validated against board-certified clinicians with high concordance, a suite of adversarial agents autonomously mutates health-related test cases to uncover vulnerabilities in real time. Applying DAS to 15 proprietary and open-source LLMs revealed a profound gap between high static benchmark performance and low dynamic reliability--the "Benchmarking Gap". Despite median MedQA accuracy exceeding 80\%, 94\% of previously correct answers failed under dynamic robustness testing. This brittleness generalized to the realistic, open-ended HealthBench dataset, where top-tier models exhibited failure rates exceeding 70\% and sharp shifts in model rankings across evaluations, suggesting that high scores on established static benchmarks may reflect superficial memorization. We observed similarly high failure rates across other domains: privacy leaks were elicited in 86\% of scenarios, cognitive-bias priming altered recommendations in 81\% of fairness tests, and hallucination rates exceeded 74\% in widely used models. By converting LLM safety evaluation for health from a static checklist into a living adversarial audit, DAS provides a scalable framework for surfacing latent risks before such systems are deployed in consumer-facing health assistants, clinician-facing tools, and broader healthcare workflows. Code is available at https://github.com/JZPeterPan/DAS-Medical-Red-Teaming-Agents.
arXiv:2503.20920v3 Announce Type: replace
Abstract: The non-Hermitian Bethe-Salpeter eigenvalue problem, in the definite case, is a structured eigenproblem, with real eigenvalues coming in pairs $\{\lambda,-\lambda\}$ where the corresponding pair of eigenvectors are closely related, and furthermore the left eigenvectors can be trivially obtained from the right ones. We exploit these properties to devise three variants of structure-preserving Lanczos eigensolvers to compute a subset of eigenvalues (those of either smallest or largest magnitude) together with their corresponding right and left eigenvectors. For this to be effective in real applications, we need to incorporate a thick-restart technique in a way that the overall computation preserves the problem structure. The new methods are validated in an implementation within the SLEPc library using several test matrices, some of them coming from the Yambo materials science code.
arXiv:2607.15195v1 Announce Type: cross
Abstract: This paper introduces a dynamic portfolio optimization framework for large institutional investors using Scientific Physics-Informed Reinforcement Learning (SciPhyRL). Formulated in continuous time over an extended state space that includes explicit cumulative costs, the approach leverages offline historical data to learn optimal, distribution-aware strategies. A core innovation reduces the optimization challenge to solving an HJB equation by projecting it onto observed trajectories as a pathwise Hamilton-Jacobi equation. This is solved directly from data using PINN in a single offline sweep, eliminating the need for traditional value or policy iteration. To make the method effective at practical short horizons, the control variable is recast from a continuous trading rate to a discrete target holding. This ensures signal-implied positions are reached immediately, while execution costs are evaluated against a microstructure-grounded quadratic price impact model. Evaluated on a $14$-asset ETF universe using an engineered oracle signal, the learned Gibbs policy yields substantial out-of-sample Sharpe ratio improvements over static and myopic baselines. The results demonstrate that the proposed framework successfully translates known signal quality into a robust, multi-period, and cost-aware allocation mechanism with strictly controlled volatility and turnover.
arXiv:2505.16839v4 Announce Type: replace
Abstract: Modern Vision-Language Models (VLMs) can solve a wide range of tasks requiring visual reasoning. In real-world scenarios, desirable properties for VLMs include fast inference and controllable generation (e.g., constraining outputs to adhere to a desired format). However, existing autoregressive (AR) VLMs like LLaVA struggle in these aspects. Discrete diffusion models (DMs) offer a promising alternative, enabling parallel decoding for faster inference and bidirectional context for controllable generation through text-infilling. While effective in language-only settings, DMs' potential for multimodal tasks is underexplored. We introduce LaViDa, a family of VLMs built on DMs. We build LaViDa by equipping DMs with a vision encoder and jointly fine-tune the combined parts for multimodal instruction following. To address challenges encountered, LaViDa incorporates novel techniques such as complementary masking for effective training, prefix KV cache for efficient inference, and timestep shifting for high-quality sampling. Experiments show that LaViDa achieves competitive or superior performance to AR VLMs on multi-modal benchmarks such as MMMU, while offering unique advantages of DMs, including flexible speed-quality tradeoff, controllability, and bidirectional reasoning. On COCO captioning, LaViDa surpasses Open-LLaVa-Next-8B by +4.1 CIDEr with 1.92x speedup. On bidirectional tasks, it achieves +59% improvement on Constrained Poem Completion. These results demonstrate LaViDa as a strong alternative to AR VLMs. Code and models will be released in the camera-ready version.
arXiv:2405.20983v3 Announce Type: replace
Abstract: Goal-oriented communication (GoC) is a form of semantic communication where the effectiveness of information transmission is measured by its impact on achieving the desired goal. In Internet-of-Things (IoT) networks, GoC can enable sensors to selectively transmit data relevant to the intended goals of the receiver, thereby facilitating timely decision-making, reducing network congestion, and enhancing spectral efficiency. In this paper, we consider an IoT scenario where an edge node polls sensors monitoring the state of a non-linear dynamic system (NLDS) to respond to the queries of several clients. This work delves into the foregoing GoC problem and solution, which we termed goal-oriented scheduling (GoS). The latter utilizes deep reinforcement learning (DRL) with meticulously devised action space, state space, and reward function. A long short-term memory network is used to estimate the inter-query duration and the corresponding estimation standard deviation. This empowers the proposed DRL scheduler to make judicious decisions, even when no queries are posed. Numerical analysis demonstrates that the proposed GoS reduces the mean square error (MSE) of the query responses compared to the benchmark scheduling methods even as the number of clients and DRL action space increase, which proves its scalability. Moreover, this is attained without polling sensors during $70\% - 87\%$ of the testing phase, thus promoting energy efficiency. Lastly, the complexity of the proposed GoS is relatively lower than the benchmark scheduling methods.
arXiv:2406.06760v3 Announce Type: replace
Abstract: Mix networks (mixnets) enable anonymous packet-based communication via multi-hop routing. Measuring the reliability of mix nodes serving as intermediate hops in a way that is decentralized, accurate, public, and compatible with low-latency packet routing, remains, however, a challenge: existing verifiable mixing schemes introduce routing delays ranging from minutes to hours that grow with traffic volume, severely limiting their practicality for large-scale or latency-sensitive applications. We propose a decentralized scheme that provides public reliability estimates for a mixnet's links and nodes without increasing latency for client traffic. The scheme achieves optimal complexity, independent of total network traffic, and enables accurate, low-overhead estimation even at scale. Our approach relies on verifiable measurement packets generated via a novel VRF-based routing primitive. This mechanism produces unforgeable measurement packets indistinguishable from normal traffic, while ensuring that all packet routing choices are consistent with the mixnet's routing policy. Reliability scores are derived from revealed measurement packets. The scheme remains robust under both unreliable and adversarial conditions, producing accurate reliability estimates as long as every honest node has a majority of honest predecessors and successors in the routing graph. We validate our design experimentally, demonstrating its practicality and effectiveness.
arXiv:2607.15148v1 Announce Type: cross
Abstract: A modular continuous aperture array (CAPA)-based multi-user communication system is investigated, where only a portion of the aperture, namely sub-CAPAs, is activated to serve users. The signal model for the proposed modular CAPA is first introduced. Based on this model, a spectral efficiency (SE) maximization problem is formulated to jointly optimize the sub-CAPA activation and beamforming, subject to constraints on the limited number of active sub-CAPAs and the total transmit power. To address the resulting mixed-integer optimization problem, a branch-and-bound (B&B)-based algorithm is first proposed for optimal sub-CAPA activation and beamforming design. After that, the spatial bandwidth of the modular CAPA under partial activation is analyzed. The analysis reveals that a modular CAPA with partial sub-CAPAs activated could achieve a maximum spatial bandwidth comparable to that of a conventional CAPA. Motivated by this insight, a low-complexity spatial bandwidth-aware sub-CAPA activation scheme is further proposed. Finally, numerical results demonstrate that i) modular CAPA architectures with partial activation can consistently achieve greater performance gains than adjacent CAPA activations; ii) the proposed B&B scheme outperforms all benchmark schemes in terms of SE; and iii) the proposed spatial bandwidth-aware scheme provides an attractive performance-complexity tradeoff compared with the proposed B&B-based algorithm.
arXiv:2607.15151v1 Announce Type: cross
Abstract: The entanglement-assisted capacity of a quantum channel admits an additive single-letter characterization, implying that joint encodings across channel uses cannot increase the ultimate communication rate. Here, we show that this additive picture does not extend to communication reliability. Specifically, we prove that the Petz-R\'enyi channel information can be strictly superadditive for every $\alpha\in[1/2,1)$, yielding a genuine multi-copy enhancement of the entanglement-assisted random-coding error exponent, even though the entanglement-assisted capacity remains additive. We establish this phenomenon analytically already for measurement channels, which are entanglement-breaking and have additive unassisted capacity. Remarkably, this strict superadditivity is witnessed by a separable, classically correlated two-copy channel-input marginal, demonstrating that no entanglement between the transmitted systems is required. Our results show that, although correlations across channel uses cannot increase the ultimate rate of entanglement-assisted communication, they can enhance its reliability.
arXiv:2605.03723v2 Announce Type: replace
Abstract: The rise of large language models (LLMs) has created an urgent need to distinguish between human-written and LLM-generated text to ensure authenticity and societal trust. Existing detectors typically provide a binary classification for an entire passage; however, this is insufficient for human--LLM co-authored text, where the objective is to localize specific segments authored by humans or LLMs. To bridge this gap, we propose algorithms to segment text into human- and LLM-authored pieces. Our key observation is that such a segmentation task is conceptually similar to classical change point detection in time-series analysis. Leveraging this analogy, we adapt change point detection to LLM-generated text detection, develop a weighted algorithm and a generalized algorithm to accommodate heterogeneous detection score variability, and establish the minimax optimality of our procedure. Empirically, we demonstrate the strong performance of our approach against a wide range of existing baselines. The python implementation of our proposal is available at https://github.com/Mamba413/DetectLLMSegmentation.