Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science

cs Watch Computer Science (sign in to use)

Friday 14 August

Social and Information Networks

Whose Posts Get Ranked: Identical-Text Exposure Gaps in Bluesky Custom Feeds

Bluesky lets users deploy custom feeds, independently operated recommendation algorithms that the platform serves alongside thousands of others. This paper investigates how evenly these feeds treat posts with the same text. To measure this, we take repeated snapshots of the Top-50 lists that 1,366 public feeds return, and we group posts with identical text, from different authors, that were created before the same list response and closely matched in age. Exposure diverges widely inside these matched sets, which span 250 feeds: in 33% of sets, one copy appears on the list while another does not. Fixed-effects regressions show that this divergence is associated with the author's history on the specific feed. Authors new to a feed receive less exposure for the same text (-0.061 in reciprocal-rank weight), while authors whose posts the feed has returned before receive more. A new author with more followers than the competing author still loses 74% of head-to-head comparisons. Media and post-type features show no detectable association after multiple-comparison correction. These results are early evidence that access to many independent feeds is not enough to give identical texts equal exposure.
Read the full abstract

Robotics

Knowledge-Data-Dual-Driven Reinforcement Learning for Autonomous Vehicle Control in Mixed Traffic

In mixed traffic, decision-making for autonomous vehicles (AVs) confronts three interrelated challenges. First, physics-based priors incorporated into reinforcement learning (RL) models fail to capture latent interactive vehicle intentions and diverse driver behaviors, limiting the proactive reasoning capabilities. Second, abrupt maneuvers by surrounding vehicles cause non-stationarity, leaving long-tail safety events under-explored. Third, hybrid action spaces destabilize unified RL training due to the different temporal scales of continuous car-following and discrete lane-changing maneuvers. To address these issues, we propose Knowledge-Data Dual-driven Reinforcement Learning (KDDRL). First, a conditional deep generative model synthesizes intention-aware future trajectories, converting passive perception into proactive predictive states. Second, a knowledge-data dual-driven paradigm operates on these predictive states, fusing probabilistic data-driven insights with physical constraints to guide safe exploration through safety-critical scenarios. Third, a coupling module compresses both intention-aware trajectories and physical constraints into compact shared embeddings. This unified representation enables asynchronous multi-timescale optimization of continuous car-following and discrete lane-changing while preserving mutual information. Evaluations on dataset-calibrated simulations demonstrate that KDDRL effectively handles intention uncertainty, accelerates training convergence, and outperforms conventional baseline methods in terms of safety, efficiency, and comfort.
Read the full abstract

All authors Jie Fang, Wei Zheng, Mengyun Xu, Eui-Jin Kim

Numerical Analysis

PDE Realization and Structure-Aware Solvers for a Compact Two-Stage Fourth-Order IMEX Method

This paper develops a PDE realization and solver framework for a compact two-stage fourth-order two-derivative implicit–explicit time discretization for stiff split problems. The key difficulty is a consistency–cost coupling: full-field temporal differentiation is required to preserve the mixed explicit–implicit interactions of the time integrator, but the same interactions widen the implicit stage operators and can make each solve substantially more expensive. We show that a second-order ADER/Cauchy–Kowalevski local evolution is sufficient for the fourth-order outer composition, establish smooth reconstruction consistency and a fully discrete error balance, and derive the leading Lie-bracket defect produced by self differentiation of the split fields. An inexact-stage analysis gives asymmetric midpoint and endpoint residual tolerances that preserve fourth-order accuracy. For the widened stages, the complete mixed action is retained matrix-free while only dominant stiff physics is approximated in the inverse. This yields quadratic and shifted preconditioners, Fourier and multilevel realizations, a semilinear reaction–diffusion reduction, and source-local elimination for relaxation systems. Exact quadratic cancellation, a diffusion-dominated Fourier estimate, and an $\varepsilon/\Delta x$ bound for Jin–Xin relaxation explain the main solver mechanisms. Numerical ablations verify the consistency and tolerance results, while a two-dimensional Brusselator study on two grids shows favorable error-versus-wall-time behavior over a useful accuracy range. A classical stiff-front benchmark also identifies the separate spatial shock–source limitation. The results support a regime-dependent efficiency claim rather than a universal speedup.
Read the full abstract

Information Retrieval

Predicting Custom-Feed Returns for New Bluesky Posts: A Prospective Study

The conventional approach to cold-start recommendation addresses new users or newly introduced items. Bluesky custom feeds create a different setting: independently operated feeds filter content from a shared public stream. In this setting, newly published posts are the cold-start objects, while the feeds serve as candidates. We propose a cold-start routing task in which a newly ingested public post is the query and all rankable feeds in the monitored panel are ranked according to whether each will subsequently return it. We build a still-evolving collect-first, label-later benchmark dataset. The collected dataset covers a fixed panel of 5,000 monitored feeds and contains 17.804 million public posts, 1.865 million observable post–feed return records, and 625,083 valid feed polls. The labels record whether a post is observed among a feed's AppView Top-50 results in at least one poll during the 24 hours after publication. The current experiments use two disjoint 24-hour test folds, each paired with a 24-hour training window and separated by a 24-hour outcome-availability gap. Evaluation is conditional on the 602,186 test posts that have at least one positive observed label and satisfy the metric eligibility criteria; these posts account for 9.04% of all 6,661,658 test posts. Across the two folds, LambdaRank achieves the best equal-fold mean values among the evaluated models: 0.7361 for capped Recall@10, 0.6127 for NDCG@10, and 0.7749 for Hit@10.
Read the full abstract

Hardware Architecture

Exploring High-Bandwidth Flash for Modern LLM Inference: Opportunities and Challenges

This work investigates the potential benefits and technical challenges of using high-bandwidth flash (HBF) for large language model (LLM) inference. HBF has gained increasing attention as a promising solution to mitigate memory-capacity bottlenecks in modern LLM-serving systems, but its benefits and challenges remain largely uninvestigated. To address this gap, we thoroughly analyze HBF-based LLM-serving systems under diverse system configurations and operating scenarios in which HBF serves as a main GPU-memory component to handle both reads and writes. Our analysis shows that, despite its limited write performance, HBF can significantly improve the batch size, throughput, and flexibility of LLM-serving systems while reducing the minimum GPU requirements, but realizing these benefits critically depends on sustaining HBM-comparable read bandwidth and requires significant endurance improvements.
Read the full abstract

All authors Dowon Son, Yonggon Park, Hyunuk Cho, Hyungkyu Ham, Onur Mutlu, Sungjin Lee, Gwangsun Kim, Jisung Park

Software Engineering

Engineering Reliable Coding Agents: Evaluating and Operating the System Around the Model

AI coding agents are commonly evaluated as models but deployed as systems. Their reliability depends not only on model capability, but on the harness, execution state, retrieval, memory and state management, permissions, review interfaces, and resource allocation. This monograph examines those boundaries and develops a framework for evaluating and operating coding agents reliably. It synthesizes 164 scholarly works, 100 practitioner records, 29 benchmark records, and 17 author-system case records through a structured multivocal review, targeted update audits, software-engineering coverage analysis, and distributed-systems evidence synthesis. Across this evidence, many apparent model failures originate elsewhere in the system, while improvements at one layer often fail to propagate to end-to-end outcomes. Evaluation and operation are treated as a dependency chain in which weaknesses in task construction, execution environments, retrieval, state management, verification, or observability can invalidate downstream conclusions. The monograph contributes a versioned catalog of 206 reliability records: 193 gated practices, including 56 developed in depth, plus 13 research leads; an evidence ledger; a framework for dependency and repair asymmetry across the agent lifecycle; measurements and failure cases from operated agent systems; runnable evaluation and reliability protocols; and five reusable agent skills with evidence maps. Together, these provide a system-level methodology for distinguishing model capability from infrastructure effects, designing defensible evaluations, and building systems that recover safely when components fail. The review is structured rather than exhaustive, evidence strength varies by topic, and results depend on workload and configuration. The methods record which search lanes were executed, which remain unexecuted, and limits on evidence-grading claims.
Read the full abstract

Machine Learning

Geometric Filtering of LLM-Generated Samples for Few-Shot Text Classification

Large language models (LLMs) can generate synthetic training data for text classification, but the quality of generated samples is heterogeneous: some fall in correct class regions of the embedding space while others land in peripheral or cross-class zones. We propose a geometric filtering framework that evaluates each LLM-generated sample by its Euclidean distance to real class examples in a sentence embedding space, selecting only geometrically consistent candidates. A soft weighting mechanism transforms filter scores into sample weights for classifier training. Evaluated across 13 datasets, 5 classifiers, 10 augmentation methods, and over 6,700 configurations, our method achieves +2.61 percentage points (pp) over SMOTE ($p<0.0001$, Cohen's $d=0.95$, 88.9% win rate). The approach generalizes to named entity recognition (+9.26pp, 100% win rate) without filter modification, and is robust across 5 LLMs from 4 providers. A key finding is that the simplest distance-based filter consistently outperforms complex multi-criteria alternatives.
Read the full abstract

Computer Vision

Attention Capture Is Not Detection: A Two-Stage Account of How Humans Miss Localized AI Image Edits

As AI-generated image edits proliferate, the platforms meant to curb the resulting disinformation treat detectability as a single, undifferentiated property: an edit either gets a warning or it does not. We show this is the wrong model. Across a controlled eye-tracking study ($N=59$, Latin-square design, four conditions crossing edit area and semantic plausibility), a mixed-effects analysis reveals that whether an edit is noticed and whether it is correctly judged as fake are dissociable stages, governed by different factors: edit area drives attention capture ($p<0.001$) while semantic plausibility drives judgment accuracy and look-but-fail-to-see (LBFS) error rates ($p<0.001$). This dissociation survives correction for multiple comparisons; a secondary interaction between the two factors does not. This two-stage account extends a long-standing distinction in visual attention research (between pre-attentive capture and effortful recognition) into the new domain of AI-edit detectability. We then test whether a generative eye-movement model can computationally operationalize the attention-capture stage: a Transformer trained to generate scanpaths tracks per-image attention with strong discriminative power (Pearson $r=0.77$–$0.82$ across held-out stimuli) and, on the harder task of predicting LBFS incidence, modestly outperforms a two-parameter linear baseline even without access to the plausibility label ($r=0.52$ vs. $r=0.48$). We report this comparison, our ablations, and our method's limitations (a single fixed train/validation split, not leave-one-subject-out) without inflation, consistent with responsibly communicating what a machine learning system can and cannot do to help curb AI-driven disinformation.
Read the full abstract

Social and Information Networks

Audience capture, selective exposure, affective assimilation or ideological sorting? Polarisation of climate politics under low media-party parallelism

Studies often attribute the polarisation of climate politics in majoritarian democracies to differing treatment of climate change by conservative and liberal media. How then does such polarisation occur in a system, where mainstream media do not systematically align with party positions? Focusing on how users in different ideological blocs posted and commented on news articles on Finnish Twitter between 2015 and 2023, we address four mechanisms: audience capture, where outlets adapt to increasingly differentiated audiences, reinforcing their divergence through feedback between audience engagement and editorial incentives; selective exposure, where users post ideologically congruent articles from the same outlets; affective assimilation, where users communicate identical content with ideology-consistent valence; and ideological sorting, where identity-driven alignment structures news-posting behaviour across multiple issues. Bayesian regression models indicate increasing divergence in outlet choice, with conservative-leaning users posting more tabloid content, in line with audience capture, and partial support for selective exposure, as within-outlet article posting becomes increasingly differentiated. Valence divergence increases only slightly over time, providing moderate support for affective assimilation, while issue-specific communities show weak alignment with blocs despite convergence over time. Our work refines theories of media effects beyond majoritarian democracies and contributes to understanding the current impasse in climate politics.
Read the full abstract

Artificial Intelligence

Joint Optimization of Memory and Computing Frequency for Energy-Efficient DNN Inference

Deep neural network (DNN) inference on mobile devices often incurs high latency and energy consumption due to limited computing and memory resources. To enable energy-efficient DNN inference, most existing studies focus on dynamic voltage and frequency scaling (DVFS) for adjusting the computing frequency, while the impact of memory frequency on the inference performance has been greatly overlooked. In this paper, we consider the impact of memory frequency and computing frequency on DNN inference time, and jointly optimize these two frequencies together with communication resources for energy-efficient DNN inference. Based on a realistic inference time model, we formulate an optimization problem to minimize the energy consumption of all mobile devices under the deadline constraint. For local inference, we derive a near-optimal closed-form solution via convex optimization, while an optimal closed-form solution for transmission power is obtained for edge inference with the given bandwidth. Furthermore, we propose a low-complexity heuristic algorithm to effectively solve the overall problem with polynomial time complexity. Simulation results based on measured data show that the proposed near-optimal solution for local inference can achieve optimal performance under strict deadline constraints, with a performance gap of up to 2.5% compared with the optimal solution. Meanwhile, our proposed algorithm significantly reduces the energy consumption of devices by up to 10.4% compared to other methods.
Read the full abstract

All authors Yunchu Han, Zhaojun Nan, Sheng Zhou, Zhisheng Niu

Computer Vision

XSA-MAD: Cross-modal Semantic Alignment for Morphing Attack Detection

Morphing attacks pose a serious threat to face recognition systems. However, existing image-based morphing attack detection (MAD) methods often generalize poorly to unseen generation techniques because they rely solely on visual cues. We propose XSA-MAD, a CLIP-based multimodal framework that explicitly models semantic inconsistencies between bona-fide and morphed faces. Morphing concepts are decomposed into four interpretable attributes, including identity, facial geometry, texture, and consistency, and are encoded as structured and attribute-aware textual representations. The image encoder is progressively aligned with this discriminative textual space, resulting in a unified semantic representation that captures generation-invariant and concept-level discrepancies between bona-fide and morph images. Experiments on MAD22 and MorDIFF, following training on SMDD, demonstrate strong generalization across diverse morphing principles. In particular, XSA-MAD achieves an equal error rate of 2.92% on GAN-based morphs and consistently outperforms existing methods under high-fidelity generative attacks.
Read the full abstract

All authors Jie Jin, Mahiro Tokumasu, Yu Makino, Masakatsu Nishigaki, Tetsushi Ohki

Information Theory

The Capacity Region of the Multiple Access Channel with Non-Signaling Assistance

The capacity region of the $K$-sender discrete memoryless multiple access channel (MAC) is fully characterized when non-signaling (NS) assistance is available to all $K$ transmitters and the receiver. It is shown to have the same form as the classical capacity region of the MAC, except that the input distribution is allowed to be arbitrarily dependent across the senders. In particular, the NS-assisted capacity region matches the natural generalization to $K$ senders of an outer bound that was previously established by Fawzi and Fermé for $K=2$ senders. Additionally, we provide examples of $K$-sender MACs where the multiplicative gain in capacity from NS-assistance is arbitrarily close to $K$. Combined with an upper bound from prior work, this establishes $K$ as the extremal value of the multiplicative gain from NS-assistance across all $K$-sender MAC settings.
Read the full abstract

Computer Vision

Face Re-morphing: Differential Morphing Attack Detection via Feature-Space Similarity Changes

Face morphing attacks pose a serious threat to face recognition systems because a single morphed document image can be matched to multiple contributors. Differential morphing attack detection (D-MAD) addresses this threat by comparing a document image with a trusted live image, but existing methods often rely on static feature differences, constituent-face reconstruction, or multi-cue fusion. This paper proposes Face Re-morphing, a D-MAD method that uses the feature-space response to an additional morphing operation as a detection cue. Given a document image and a trusted live image, the proposed method generates a re-morphed image and uses the change between the document–live and live–re-morphed cosine similarities as the detection score. Experiments on FRLL-Morphs and FEI Morph show that the proposed cue is effective across different morphing conditions, re-morphing methods, and face recognition models. Comparisons with existing methods show favorable results on AMSL and indicate that the proposed method performs well under the Criminal condition on FEI Morph Version~1, particularly when using MorDIFF. These results indicate that re-morphing-induced similarity change provides a complementary cue for D-MAD.
Read the full abstract

Computer Vision · Cross-listed from eess.IV

From crown candidates to neighborhood screening: integrating optical GeoAI and spatial modeling for urban-canopy assessment in Davis, California

Timely urban-canopy information is essential for linking remote sensing with heat, mobility, and neighborhood planning. We developed an optical GeoAI workflow for Davis, California, using 2022 National Agriculture Imagery Program imagery (0.6 m RGB+NIR). DeepForest generated crown candidates; an NDVI threshold, non-maximum suppression, and box-prompted Segment Anything Model (ViT-B) produced a crown-anchored canopy surface. Analyses used the 25.92 km2 Census TIGER municipal boundary and a 100 m grid. The workflow retained 11,741 candidate crowns and mapped 7.71 km2 of canopy (29.8% of the city). On the identical extent, pixel precision was 0.804, pixel recall was 0.873, and 97.4% of candidate centers agreed with the 2022 USDA/CAL FIRE LiDAR-assisted canopy product (IoU 0.719; Dice 0.837; area recovery 108.5%). Approximately 49% of candidates occurred within 15 m of a road. Canopy was inversely associated with Landsat land-surface temperature (Spearman rho = -0.477; partial rho = -0.551 controlling for built probability), and spatial-lag modeling confirmed clear neighborhood structure. Two transparent attention surfaces combined canopy need with thermal and contextual indicators. The framework provides a reproducible, updateable screening layer that complements structural canopy products and municipal inventories while retaining assumptions, data provenance, and spatial diagnostics for planning interpretation.
Read the full abstract

Computation and Language

Bootstrapping Niche Multilingual Code Translation via Reinforcement Learning with Execution-Based Verifiable Supervision

Code translation must preserve executable behavior across many programming languages, yet neural code translation has largely focused on a few popular languages such as C++, Java, and Python. This leaves a niche, many-to-many setting where parallel supervision is sparse, producing plausible but non-executable translations. We address this setting with preference-based reinforcement learning driven by execution-based supervision. Our pipeline firstly expands verifiable seed Python programs into a multilingual pool of execution-validated codes. Using the pool, a base LLM generates translation candidates across language pairs, which we label by their execution outcomes. The resulting preferences are used to train a reward model that scores cross-language translation quality. Finally, we optimize our base LLMs with GRPO over 600 directed language pairs (25 x 24) using the reward model as a signal. To evaluate the niche translation capability, we introduce HumanEval-X++, an execution-based benchmark that extends HumanEval-X to a broad many-to-many language space. We evaluate our approach using Qwen-3.5 4B and 9B models. On HumanEval-X++ and existing benchmarks, it yields consistent gains over the untrained baselines. In particular, the 4B model achieves an average improvement of 13% across all languages on HumanEval-X++, with a gain of 21% on mid-tier languages. Our study establishes a reliable approach of data generation, training, and benchmarking, paving the way toward further bootstrapping the quality of many-to-many translation for programming languages.
Read the full abstract

All authors Kouki Yuki, Jie Zeng, Kyoko Ogawa, Ryunosuke Ikeda, Yohei Kobashi, Takeshi Kojima, Ikuya Yamada, Yusuke Iwasawa, Yutaka Matsuo

Cryptography and Security

Verified Pythagorean Composition for Adaptive Cryptographic Games: Noise Flooding in Homomorphic Encryption

Noise flooding is a standard defense against decryption attacks on approximate homomorphic encryption, but its security proof is unusually sensitive to composition. Replacing each of $q$ adaptive decryption answers with a statistically close simulation and applying an ordinary hybrid argument loses linearly in $q$. The cryptographic proof instead accumulates conditional Kullback-Leibler (KL) costs and converts to statistical distance once, giving the parameter-critical square-root loss. We machine-check this argument using Rocq and SSProve. Given any fully homomorphic encryption scheme that is approximately correct and IND-CPA secure, we formalize a reduction for every $q$-query IND-CPAD adversary and prove \[ \Pr[\mathsf{IND\text{-}CPAD}_{\mathsf{NF}}^{\mathcal A}=1] \leq \beta_{\mathsf{CPA}}(\mathcal B_{\mathcal A,q}) + \frac{\sqrt{qn}}{2\gamma}. \] where $n$ is the plaintext dimension and $\gamma$ is the flooding-width multiplier. Our proof constructs a new relational program logic over SSProve semantics. Its Pythagorean judgment composes conditional KL budgets without converting them to statistical distance, and a verified trace compiler lifts a local oracle rule to arbitrary adaptive programs with a single final conversion.
Read the full abstract

All authors Yi Lee, Alexandru Cojocaru, Junyi Liu, Xiaodi Wu

Machine Learning

Federated Prompt Learning: A Unified Framework, Empirical Analysis, and Future Directions

Large language models (LLMs) have become core components of cloud-based intelligent services in academia and industry, yet their training and deployment are hindered by high computational costs, data centralization, and privacy concerns. Federated learning (FL) offers a decentralized training paradigm that enables clients to collaboratively train a learning model without sharing raw data, making it a promising solution for privacy-preserving LLM training and reasoning. This paper presents a comprehensive survey of federated prompt learning (FPL) to review recent advances in integrating the federated learning paradigm and large language models, answering the following research questions: RQ1: The fundamental motivations, characteristics, and enabling technologies of FPL, and how it differs from conventional FL and full-model federated fine-tuning; RQ2: The trade-offs FPL approaches exhibit in performance, communication efficiency, computational overhead, scalability, personalization, and heterogeneity handling; RQ3: The remaining security, privacy, robustness, and system challenges, along with key future research directions. To this end, we systematically examine existing FPL methods across the full model lifecycle: pre-training, fine-tuning, and practical applications, while discussing security, privacy, and robustness issues and summarizing existing defense mechanisms. Finally, we highlight open challenges and future directions, aiming to help readers understand how the insights drive research in FPL.
Read the full abstract

All authors Qinglin Yang, Chen Qiu, Hongyuan Zhang, Pengdeng Li, Yuan Liu, Zhihong Tian

Sound

The MPB Corpus: A Dataset of Melody, Rhythm, Harmony, and Melody-Harmony Relationships in Brazilian Popular Music

PLEASE CHECK AND REFER TO THE PUBLISHED VERSION OF THIS PAPER ON https://doi.org/10.5334/tismir.357 This paper presents the MPB Corpus, a dataset of Brazilian Popular Music (Musica Popular Brasileira, MPB) comprising derived symbolic annotations for 500 pieces across four musical parameters – melodic contour, melodic rhythm, harmony, and the melody-harmony relationship – making it the most comprehensive dataset to date for computational musicology of Brazilian music. It is released under a CC-BY-4.0 license in an open, machine-readable tabular format, with documentation and reproducible analysis code. Because the annotations were derived from copyrighted musical scores, the corpus contains no audio recordings or score-level representations such as MIDI or MusicXML; it nonetheless provides a reusable resource for MIR research on a culturally significant repertoire that remains underrepresented in existing datasets. To support the encoding process, we introduce analytical models for capturing rhythmic and melodic information, along with tailored visualizations and summary metrics. Finally, we present a brief exploratory analysis illustrating the dataset's potential for formulating and addressing musicological questions about the genre.
Read the full abstract

Systems and Control

Real-Time In-Domain Congestion Control for the LWR Traffic Model via Control Barrier Functions

This paper presents a control barrier function-based method for real-time in-domain congestion control of the Lighthill-Whitham-Richards traffic model. Traffic congestion is formulated as a distributed safety control problem, leading to a infinite-dimensional optimization problem. Through the discretization and Karush-Kuhn-Tucker (KKT) analysis, the problem is converted into a high-dimensional quadratic program. A structure-exploiting primal-dual active set algorithm is then developed to compute the safe control input in real time, with convergence guarantees. Numerical simulations with different nominal controllers demonstrate the effectiveness and real-time feasibility of the proposed approach.
Read the full abstract

Computation and Language

ASSERT: A Measurement Pipeline for GenAI Audits

Audits of generative AI (GenAI) systems often summarize behavior as a reported rate: how often the audited system complies with policy. Researchers and stakeholders use that rate to compare systems, track regressions, and gate deployment. A reported rate reflects both the system under audit and the measurement choices behind it, so a change in the rate can leave it unclear whether the system or those choices moved. We introduce ASSERT, a specification-driven measurement pipeline for GenAI audits that ties each reported rate to a written specification of the measurement choices used to produce it. ASSERT helps draft a behavioral rubric and test cases, then runs the audit against a GenAI system and returns a reported rate. In a case study on conversational deception, we observe that the reported rate moves substantially with the dialogue setup, the simulated user, the judge, and the evidence bar for non-compliance. These measurement choices substantially change the reported rate and can reorder GenAI system rankings. Because each reported rate is tied to an explicit specification, differences across audits are easier to attribute and interpret.
Read the full abstract

All authors Riccardo Fogliato, Abhinav Palia, Xiawei Wang, Emily Sheng, Chad Atalla, Jean Garcia-Gathright, Nicholas Pangakis, Sharman Tan, Dan Vann, Hannah Washington, P. Alex Dow, Heba Elfardy, Hanna Wallach, Sandeep Atluri

Numerical Analysis

Vanishing Clutter: Fast and Accurate Shape Imaging via Active Cloaking

Imaging a target sample in near-field scanning optical microscopy (NSOM) is fundamentally limited by measurement artifacts and data contamination from multiple probe-sample scattering. The probe, essential for subwavelength resolution, inherently perturbs the local field, degrading the signal-to-noise ratio and rendering the inverse problem for quantitative shape reconstruction highly ill-posed. We first establish the well-posedness of the corresponding forward model, providing a rigorous foundation for subsequent imaging. We then repurpose active cloaking–conventionally the antagonist of imaging–as an enabling mechanism to eliminate probe-induced interference. Rather than directly reconstructing the sample from corrupted data, we actively cloak the probe by formulating an optimal control problem and prove the existence and stability of its minimizers. Leveraging the theory of localized anomalous resonance in layered plasmonic structures, we derive an exact closed-form minimizer, thereby circumventing the computationally prohibitive iterative solution of the optimal control problem. The resulting cloaking-driven interference removal yields a virtually probe-free measurement environment, enabling fast, artifact-free shape reconstruction. Extensive numerical experiments demonstrate accurate shape reconstruction and dramatic acceleration–often by orders of magnitude–over conventional iterative methods applied directly to probe–contaminated data without cloaking–based preprocessing, validating the robustness and transformative potential of the proposed approach for high-fidelity subwavelength imaging.
Read the full abstract

Thursday 13 August

Computers and Society

The EU AI Act Beyond Europe: A Practical Readiness Framework for Non-EU Organisations. Comparative Evidence from Four African Countries

The EU Artificial Intelligence Act has implications for organizations within Europe as well as organisations outside the European Union where they provide AI-enabled services to EU clients, use European vendors, participate in EU-facing supply chains, or generate outputs that are used in the EU. This study examines the implications of these impacts for African countries and other non-EU organisations, using South Africa, Nigeria, Kenya and Egypt as comparative cases. It applies doctrinal legal analysis, structured document review and scenario-based comparison to assess the sufficiency or otherwise of existing regulatory frameworks within the selected countries. The analysis shows that the existing regulatory frameworks (like data-protection, financial-sector, cybersecurity and corporate-governance rules) can support AI readiness, but they do not resolve questions of territorial scope, role classification, risk categorisation, vendor responsibility and access to compliance evidence. The analysis proposes a seven-step readiness framework which covers exposure mapping, AI inventories, role and risk classification, governance, contracts, evidence management and continuous monitoring. It argues for proportionate, interoperable approaches that enable non-EU organisations to demonstrate readiness without treating EU compliance as a uniform exercise.
Read the full abstract

Human-Computer Interaction

AI-Powered Symptom Assessment and User Experience: A Case Study of Simtomi and Simtomi-Care

Digital symptom checkers are widely used for quick guidance on health concerns, yet many systems still face challenges in collecting accurate information, supporting communication, or integrating with clinical workflows. To explore how these tools function in real use, we examine the case of the Simtomi system, which pairs a multilingual symptom assessment application with a provider-facing platform. Empirical studies were conducted in two countries. In South Korea, based on participants' firsthand experience, we found that the system improved how patients communicated their symptoms and helped clinicians review cases more efficiently through structured summaries aligned with diagnostic reasoning. In the United States, responses from prospective users and healthcare professionals highlighted the value of multilingual support, structured questioning, and the system's potential to assist clinical coordination. These findings offer a grounded account of how AI-based symptom assessment tools can operate across different healthcare contexts and provide broader insight into usability, trust, and usefulness in digital health.
Read the full abstract

All authors Jinha Lee, Chan Hyung Lee, Hyunsung Lee, Seunghwan Kim, Ban Hyung Lee, Minjun Shin, Hojin Shin, Jungdo Park

Computers and Society

HIPAA-Compliant AI Deployment Patterns in Clinical Settings: Privacy-Preserving Techniques and Governance Controls

Artificial intelligence (AI) is moving from research prototypes into clinical workflows, yet the deployment of AI systems that process protected health information (PHI) remains constrained by the U.S. Health Insurance Portability and Accountability Act (HIPAA) and by the absence of shared engineering guidance for satisfying it. Existing work tends to treat privacy-preserving machine learning and regulatory governance as separate concerns, leaving practitioners without a concrete mapping from architectural decisions to compliance obligations. This paper contributes a catalog of five reusable deployment patterns for clinical AI that jointly address the HIPAA Privacy and Security Rules: a de-identified analytics zone, cross-entity federated learning, confidential inference within trusted execution environments, a PHI-minimizing retrieval-augmented clinical assistant, and a synthetic data sandbox for development. Each pattern is specified in terms of context, governing forces, structure, and residual risk, and is mapped to the administrative, physical, and technical safeguards of the HIPAA Security Rule as well as to the NIST AI Risk Management Framework and ISO/IEC~42001. We further present a threat model for clinical AI and a decision procedure for selecting patterns based on data sensitivity, trust boundaries, and latency requirements. The goal is to give security engineers, compliance owners, and clinical informaticists a common vocabulary that turns abstract regulatory duties into verifiable architectural controls.
Read the full abstract

Logic in Computer Science · Cross-listed from math.AG

Projective blowups: a formal and multicentered proof

We give a machine-checked proof, in Lean 4, of the universal property of multi- centered blowups: given a finite family Z of closed subschemes of a scheme X, presented on each chart by ideals, there is a scheme BlZ X over X, terminal among X-schemes on which every member of the family becomes an effective Cartier divisor. To the best of our knowledge this is the first formalisation of blowups, single- or multicentered, in any theorem prover. The proof is organised around the universal property of the multicentered dilatation of a ring. The comparison isomorphisms on overlaps, the cocycle identity, and the independence of the chosen presentation are obtained from the uniqueness part of this universal property. We give the exact theorem statement as formalized, the proof architecture, and the main new mathematical ingredients, together with the corresponding Lean code.
Read the full abstract

Game Theory

A Statistical Analysis of Three Player Auction Bridge

Three-player Auction Bridge is a finite imperfect-information game in which dynamic partnerships create a distinctive interaction between scoring, strategic incentives, and payoff distribution. This paper develops a unified statistical and game-theoretic framework to evaluate a traditional scoring rule against a modified mechanism designed to improve strategic incentives. We first identify a structural defect in the original scheme: a bid-invariant slam bonus can make lower contracts strictly more attractive than higher ones. A bid-dependent correction removes this incentive distortion. We then compare the two schemes using higher-moment analysis, fairness measures, Nash equilibrium analysis, and General-Sum Counterfactual Regret Minimization (GS-CFR). The results show significant differences in the shape of the payoff distributions and reveal that, although both schemes remain highly balanced across physical seats, the modified scheme substantially increases inequality across strategic roles. The game is shown to be neither zero-sum nor constant-sum, motivating a general-sum rather than minimax analysis. Full-game GS-CFR further indicates a substantial increase in the bidder's payoff under the modified scheme, accompanied by a reduction in defender payoff and a shift from separating to partially pooling bidding behavior. These findings demonstrate that scoring-rule design can fundamentally reshape incentives, payoff distribution, and information transmission in imperfect-information games. The proposed framework provides a systematic approach to evaluating such mechanisms from both statistical and strategic perspectives.
Read the full abstract

Artificial Intelligence

Coding Agents are Strong Prompt Optimizers

Search-based prompt optimizers improve prompts through iterative search: they propose edits, execute fresh rollouts, score the resulting trajectories, and retain only edits that improve a validation metric. We show that this optimization loop is unnecessary. Given only a static corpus of agent trajectories, an off-the-shelf coding agent can directly synthesize an optimized prompt, requiring neither environment access nor validation data. We call this approach Coding-Agent Skill Distillation (CASD). The key insight is reflection scope. Rather than reasoning over a small batch of trajectories at each optimization step, the coding agent writes and executes analysis code to compute corpus-wide statistics, identifies systematic failure modes, inspects representative episodes, and distills the resulting insights into behavioral rules. Across four agentic benchmarks (ALFWorld, $\tau^2$-bench retail and telecom, and SpreadsheetBench-Verified), under matched data access, a single CASD pass outperforms GEPA, a state-of-the-art reflective prompt optimizer, on three of four benchmarks and outperforms validation-gated reflective search (SkillOpt) on all four, improving the unoptimized baseline by 16.6 percentage points on average versus 10.9 for GEPA and 5.3 for SkillOpt. Because CASD performs a single offline analysis pass rather than iterative search, producing an optimized prompt costs approximately \$1.60---over $22\times$ cheaper than validation-gated search. Even when competing methods are granted additional validation data and unrestricted environment access, CASD remains ahead on two of four benchmarks. These results suggest that corpus-scale statistical reflection is a viable alternative to iterative search for prompt optimization.
Read the full abstract

All authors Agamdeep Singh, Srishti Gautam, Priyanshu Gupta, Nikita Mehrotra, Tanmay Bakshi, Sumit Gulwani

Data Structures and Algorithms

Online Line Aggregation with Deadlines: Randomized Guarantees and Learning-Augmented Tradeoffs

We study online line aggregation with deadlines, where requests arrive over time on the positive half-line and a service at location $y$ clears all pending requests in $[0,y]$ at cost $y$. In the classical adversarial setting, we propose an $e$-competitive randomized algorithm against an oblivious adversary and prove a matching lower bound. Thus $e$ is the optimal randomized competitive ratio. We then consider advice in the form of an offline feasible solution. For every confidence parameter $\lambda\in(0,1]$, our deterministic learning-augmented algorithm is $(1+3/\lambda)$-robust and $(1+3\lambda)$-consistent. We also propose a randomized learning-augmented algorithm that is $(e+e/\lambda)$-robust and $(e-1+\lambda)$-consistent against an oblivious adversary. For the offline problem, we present a polynomial-time dynamic programming algorithm. Numerical experiments complement the worst-case analysis: accurate advice lowers service costs, while both learning-augmented algorithms remain stable as the advice becomes increasingly noisy.
Read the full abstract

Cryptography and Security

CRT-Decomposed $\Sigma$-Protocols for CSIDH

We construct a zero-knowledge proof of knowledge for the CSIDH group action that exploits the Chinese Remainder Theorem (CRT) structure of the ideal class group, available whenever the group structure is known exactly, as for CSIDH-512. We prove perfect completeness, perfect special honest-verifier zero-knowledge, and 2-special soundness, in which the secret is recovered from two accepting transcripts by one subtraction and one modular inversion per CRT component; no rewinding loop, lattice reduction, or heuristic sampling appears in the extractor. Because each round admits exactly two responses, Unruh's transform yields a non-interactive proof with straight-line extraction in the quantum random oracle model (QROM), removing the multiplicative forking-lemma loss. We instantiate the scheme on the CSIDH-512 class group, verify the algebraic layer by machine over the exact 258-bit modulus ($10^4$ random instances, all passing), and report protocol-level simulations in the exponent model: Monte Carlo soundness rates matching the proven $2^{-t}$ bound within 95% confidence at every tested $t$, serialized signature sizes within 1.6% of the formulas, instrumented action counts, and a scaled meet-in-the-middle attack whose measured cost follows the predicted $\sqrt{q}$ law. We further prove two delimiting results: CRT decomposition cannot enlarge the per-round challenge space, and publishing the CRT hop curves lowers classical key-recovery cost from about $2^{128.6}$ to about $2^{67.3}$ group action evaluations. The construction is therefore correct and structurally complete today, but quantitatively secure only on future parameters whose class number has large prime factors. We compare against CSI-FiSh, CSI-Otter, and Tanuki; a tightly secure blind signature from this proof of knowledge is left to future work.
Read the full abstract

Machine Learning

Information-Theoretic Decoupled Prompt Tuning for Continual Learning

Continual learning (CL) aims to incrementally acquire knowledge from sequential data while avoiding catastrophic forgetting. Recently, prompt tuning has attracted increasing attention as an efficient approach for adapting pre-trained models to CL tasks. However, existing prompt design paradigms commonly suffer from retrieval dependence and classifier bias, which make model adaptation sensitive to prompt selection and bias predictions toward newly arrived classes. To address these challenges, we propose Decoupled Prompt Tuning for Continual Learning (DPT4CL), which decouples the CLIP textual prompt into a task-shared prompt distribution and class-specific prompts. The task-shared prompt distribution is derived by optimizing an Information Bottleneck objective to facilitate cross-task knowledge transfer and alleviate classifier bias, while class-specific prompts enhance inter-class separability without relying on explicit prompt retrieval. Furthermore, we establish a unified excess risk bound from an information-theoretic perspective, providing theoretical support for the robust generalization and forgetting mitigation of the proposed framework. Extensive experiments on standard CL benchmarks demonstrate that DPT4CL achieves state-of-the-art performance. The source code is available at https://github.com/Cloudfly-Z/DPT4CL
Read the full abstract

All authors Yunfei Zhang, Wen Wen, Tieliang Gong, Weizhan Zhang