Peer Revenue Fully funded

Papers

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

Computer Science

cs Watch Computer Science

Friday 21 August

Performance

Portable to Efficient: Auto-Tuning Hardware-Agnostic GPU Kernels in Julia

Floris-Jan Willemsen, Evelyne Ringoot and Alan Edelman

Traditionally, GPU kernels have been developed and optimized within vendor-specific programming models to achieve high performance, resulting in software that is difficult to optimize and adapt across increasingly heterogeneous computing systems. Hardware-agnostic programming models offer a more sustainable approach to GPU software development by improving portability and maintainability, but achieving efficient execution across diverse architectures remains challenging. We address this challenge by integrating auto-tuning into hardware-agnostic GPU kernels written in Julia. We rebuild the established Kernel Tuner auto-tuning framework with Julia support, enabling systematic exploration of kernel configurations for hardware-agnostic GPU kernels targeting NVIDIA, AMD, Intel, and Apple GPUs. We demonstrate this approach on hardware-agnostic singular value decomposition (SVD) as implemented in the NextLA.jl linear algebra library. The results show that auto-tuning is essential for creating resource-efficient hardware-agnostic GPU kernels across a variety of hardware. Optimal configurations improve kernel performance by a factor of 3x to 7x compared to median parameter configurations, demonstrating the substantial impact of tuning on efficient hardware utilization.
Read the full abstract

Artificial Intelligence

Ontology-supported AI Model and Dataset Management

Jan Novacek, Ali Ahari, Tobias Müller et al. (+3 more)

Recently, there has been a great deal of research into improving AI methods and their application. The main focus is on tracking progress, enabling transparent comparisons, and fostering a more profound understanding of AI. In that process, different organizations generate and use plenty of assets that need to be tracked, traced and managed. Moreover, it is important to discover assets relevant for the task at hand. This paper presents research aiming to contribute to answering the question of what is required to exchange and manage AI models and related assets effectively without semantic gaps in an industrial context. We introduce a platform for AI model exchange, which facilitates the usage, exchange, and analysis of AI models and datasets. The platform incorporates an ontology that can foster a more profound common understanding of what is required in these tasks and help tackle the issues mentioned above. Finally, we elucidate the utility of the platform through the illustration of a use case in the context of real-time critical systems.
Read the full abstract

All authors Jan Novacek, Ali Ahari, Tobias Müller, Sebastian Reiter, Alexander Viehl, Oliver Bringmann

Hardware Architecture

Event-triggered Implicit Perturbation for Zeroth-Order Fine-Tuning of Spiking Transformers

Tengteng Lei, Prabodh Katti, Rashi Dutt et al. (+5 more)

Zeroth-order (ZO) optimization estimates gradients using only forward-pass evaluations, making it suitable for fine-tuning non-differentiable, event-driven spiking neural networks (SNNs). However, its deployment on in-memory computing (IMC) accelerators is constrained by the repeated read-modify-write (RMW) operations arising from explicit weight perturbation and the prohibitive hardware footprint of random number generators (RNGs) for statistically independent per-weight perturbations. To address these challenges, we propose an implicit-perturbation ZO (IPZO) architecture in which perturbation sums computed by an event-triggered perturbation generation unit (PGU) are combined with the weighted sums produced by the IMC array, eliminating perturbation-induced RMW operations while preserving weight-stationary execution of IMC. By exploiting spike sparsity, the PGU generates and accumulates perturbation contributions only for spike-activated weight rows, reducing the required row dimension of the RNG array. An address-driven XOR recombination scheme (PGU-XOR) is further introduced to mitigate the spatial correlations caused by direct RNG reuse (PGU-Reuse). The results show that (1) PGU-XOR matches software RNGs in accuracy on Spikingformer/CIFAR-10 (76.41% vs. 76.53%) and perplexity (PPL) on SpikeGPT/WikiText-2 (54.20 vs. 53.23), whereas PGU-Reuse degrades accuracy by 9.56 percentage points and increases PPL by 11.8; (2) implemented in a TSMC 16-nm CMOS technology, PGU-XOR incurs 40.3%-46.0% area and 15.2%-48.9% energy overhead per matrix-vector multiplication relative to PGU-Reuse, yet its faster convergence reduces the total perturbation energy to 0.51x that of PGU-Reuse at iso-accuracy; (3) IPZO reduces the perturbation energy to 0.46x-0.83x that of conventional explicit weight perturbation for a batch size of B=64 and T=4 time steps, with the advantage growing as BT decreases.
Read the full abstract

All authors Tengteng Lei, Prabodh Katti, Rashi Dutt, Houssem Sifaou, Tan Peng, Osvaldo Simeone, Kai Xu, Bipin Rajendran

Software Engineering

The Substitution Escrow Threshold: When "Compatible With" Becomes Safe Enough to Buy

Amadeus Brandes

Enterprise infrastructure buyers routinely evaluate compatibility claims–"S3-compatible," "PostgreSQL-compatible," "OpenAI compatible"–as proxies for future substitution options. Yet most compatibility claims do not escrow the substitution path they imply. This paper introduces the Substitution Escrow Threshold, a five-condition framework that determines when a compatibility claim genuinely reduces institutional risk versus merely reducing first-integration cost. The five conditions–boundary closure, executable conformance, custody independence, state and operations reversibility, and extension quarantine–are applied to five infrastructure cases (OCI, Kubernetes, OpenTelemetry, S3, PostgreSQL) that populate five distinct outcome cells. The framework produces actionable diagnostics for enterprise architects, platform engineers, procurement teams, and investors evaluating compatibility-dependent infrastructure decisions, and identifies AI infrastructure as the framework's most urgent next application.
Read the full abstract

Human-Computer Interaction

Who Trusts AI with Their Emotions? Trust Formation and Sociodemographic Variation in LLM Use for Emotional Support

Natalia Amat-Lefort, Mert Yazan, Amanda Cercas Curry et al. (+1 more)

Trust in AI for emotional support is not universal; it is shaped by who users are, where they come from, and what they value. Yet research in this area lacks validated psychometric instruments for assessing user perceptions in affective AI contexts and large-scale evidence on how trust formation varies across user segments. To address these gaps, we develop and validate a seven-construct psychometric scale, test a Structural Equation Model (SEM) linking system attributes to Trust and Perceived Benefits as mediators of Actual System Use, and conduct a Multi-Group Analysis (MGA) across five sociodemographic dimensions (gender, age, education, socioeconomic status, cross-national region), drawing on 1,343 active users from seven countries. We find that users experience empathy and anthropomorphism as a unified "Humanlikeness" construct, and that Privacy, Personalization, and Humanlikeness drive Trust while Perceived Bias degrades it. Notably, adoption logic diverges across groups: Privacy shapes women's trust more than men's, Anglosphere (UK, USA) users respond more positively to Humanlikeness than Europeans, and educated and higher-income users require Trust to engage, whereas older adults and lower socioeconomic groups bypass it entirely, relying on perceived practical benefits (e.g., 24/7 availability, non-judgmental support). Our findings extend technology acceptance theory and inform the equitable design of emotional support AI.
Read the full abstract

All authors Natalia Amat-Lefort, Mert Yazan, Amanda Cercas Curry, Flor Miriam Plaza-del-Arco

Artificial Intelligence

Enhancing LLMs in Predictive Political QA with Semi-Structured Data

Yinan Liu, Zihan Zhou, Zichun Jin et al. (+3 more)

Predictive political question answering (QA), such as predicting how a political actor will vote, goes beyond factual lookup. External political resources offer rich historical evidence, but rarely contain the answer itself. Existing LLM augmentation methods, including actor-profile-based simulation and knowledge graph evidence injection, improve political reasoning but largely treat external resources as knowledge-based evidence, leaving prediction-relevant signals under-modeled. We identify two complementary signals for predictive political QA: actor stances that capture issue-specific preferences, and high-order structure signals that capture indirect dependencies among political actors. We propose PSL, a dual-view framework that converts semi-structured political records into inference-oriented evidence for LLMs. PSL extracts stance signals from question-relevant actor records in a semantic view, and learns structure-aware actor representations from an actor interaction graph in a vector view. Across three real-world datasets and multiple LLMs, PSL consistently outperforms baselines, with ablations confirming the complementary gains of stance and structure signals.
Read the full abstract

All authors Yinan Liu, Zihan Zhou, Zichun Jin, Xinyu Wang, Bin Wang, Xiaochun Yang

Artificial Intelligence

Personalized Privacy Control in LLMs via Attention Head Intervention

Junseok Kim, Nakyeong Yang and Kyomin Jung

The rise of agentic AI enables LLMs to access diverse user data, raising critical privacy concerns. Prior work on contextual privacy studies whether LLMs regulate information disclosure according to context-dependent norms. However, acceptable disclosure boundaries may vary across users even within the same context. To address this limitation, we introduce personalized privacy, which incorporates user-specific disclosure preferences into privacy control. We further present P3Bench~(Personalized Privacy Preservation Benchmark), a novel benchmark extending contextual privacy policies with personalized disclosure policies. Experiments show that prompt-based policies fail to reliably enforce personalized privacy policies, with Qwen2.5-7B and Gemma3-4B showing average policy ignorance ratios of 51.25% and 74.28%, respectively. Finally, to address this problem, we propose Repair, a robust inference-time attention head intervention method that adjusts disclosure behavior toward policy-consistent responses. Our method significantly improves adherence to user-specific privacy preferences by reducing cases where the model fails to follow the given policy.
Read the full abstract

Software Engineering

Specification Portability Across LLM Development Agents: Cross-Agent Compatibility in Specification-Driven Software Migration

Oleg Grynets, Oleksii Ilchuk, Dariia Zatulna et al. (+1 more)

This paper investigates cross-agent specification portability using Oracle-to-PostgreSQL migration as a controlled software transformation task. The study combines two experimental stages. First, a specification-first migration pipeline was evaluated on 1,006 PL/SQL files, of which 623 were successfully regenerated and 380 generated scripts executed successfully in PostgreSQL 16. Second, cross-agent experiments were conducted on a dataset of 1,802 Oracle scripts with corresponding PostgreSQL implementations using Amazon Kiro, Google Gemini, and GitHub Copilot, with Claude Code and Cursor included in the initial single-agent evaluation. Native and foreign specifications were assessed using Token F1, exact match, SQL syntax validity, AST exact match, AST mean similarity, and immediate runnability. The results show that specification size alone does not predict implementation quality and that cross-agent transfer can produce substantial agent-dependent degradation. The strongest replicated case occurred when Gemini directly consumed a Kiro-origin specification, producing a Token F1 of 0.035, SQL syntax validity of 2.33%, and AST mean similarity of 0.015. Rewriting substantially improved Gemini in the tested configuration, compression did not provide a universal benefit, and retrieval-augmented ingestion was the only common strategy represented on the per-agent Pareto frontiers of both Gemini and Copilot. The findings suggest that specifications in heterogeneous SDD workflows should not automatically be treated as agent-neutral artifacts and motivate explicit consideration of specification portability, agent-specific interpretation, and retrieval-based access in multi-agent software engineering.
Read the full abstract

All authors Oleg Grynets, Oleksii Ilchuk, Dariia Zatulna, Vasyl Lyashkevych

Machine Learning

Curriculum-Aware Interpolate-then-Refine: Learned Physiological Time-Series Imputation under Realistic Missingness

Yu-Chao Huang, Haochen Zhang, Nicholas Konz et al. (+1 more)

Imputing physiological time series (arterial blood pressure, blood glucose, etc.) is essential for addressing the missingness that pervades clinical data. Yet modern imputation methods perform poorly in this domain: a recent benchmark found that simple linear interpolation outperformed every learned imputer on real-world clinical signals with realistic gaps. We show that this reflects two properties of physiological missingness that generic imputers ignore: gaps may occur when the signal is clinically extreme rather than typical, and gap lengths can easily span orders of magnitude. To this end, we introduce Curriculum-Aware Interpolate-then-Refine (CAIR), a two-stage framework for physiological time-series imputation. Our key motivation is to learn a coarse base curve and then repeatedly correct it toward physiological realism, rather than predict a gap in a single pass. Consequently, CAIR couples a bidirectional-GRU interpolator with a Transformer refiner that corrects its own estimate over three successive passes, trained jointly under a broad, signal-agnostic random-gap curriculum. We evaluate imputers stratified by gap length and missingness mechanism (MCAR, MAR, NMAR) rather than by a single average, and CAIR is the most accurate under every mechanism on continuous glucose monitoring (AI-READI) and arterial pressure in intensive care (MIMIC-III). Its margin over the strongest baseline grows with difficulty, from 9% under MCAR to 19% under value-dependent dropout, where generic learned imputers are weakest. We further show low reconstruction error alone does not recover the burden metrics clinicians act on: interpolants matching CAIR's error fail to preserve those metrics, imputers that recover them are far less accurate, and CAIR alone ranks among the best on both axes.
Read the full abstract

All authors Yu-Chao Huang, Haochen Zhang, Nicholas Konz, Tianlong Chen

Computation and Language

No PUN Intended: Plausible Unknown Names for Person-Centred LLM Evaluation

Dimitri Staufer, David Hartmann and Ibrahim Baroud

Person names are widely used as prompt variables in LLM evaluations of factuality, privacy leakage, bias and abstention, but when a name's evidential status is uncontrolled, measurements may conflate memorisation, retrieval, name priors and wrong-person attribution. We operationalise an unknown name as one with plausible First-Last form, no indexed full-name evidence, and no ambiguity signals under a documented validation run, and introduce PUN (Plausible Unknown Names), a protocol for constructing and validating such names, combining Wikidata-derived components, web-enabled LLM screening, and controlled search revalidation. We report acceptance rate, reproducibility, ablations, and a 204-participant human study, finding accepted names are more name-like than controls while participants recover person evidence in only 3% of cases. We release 300 names with comparison controls.
Read the full abstract

Cryptography and Security

Workplace Surveillance and Insider Threat Risk Management: Legal Limits and Privacy Harms

Haywood Gelman, John D. Hastings, Suvineetha Herath et al. (+1 more)

Workplace surveillance is used by organizations to protect corporate assets and monitor employee productivity. This research presents two central arguments on workplace surveillance: although surveillance serves legitimate organizational purposes, over-surveillance can violate legal requirements and data privacy principles; and a primary security objective of workplace surveillance is the detection of insider threats (InT). InT are comprised of individuals with authorized resource access whose intentional or unintentional actions may damage or compromise corporate assets. This paper investigates InT personas to understand behavioral and psychological detection criteria. Employee surveillance tools and techniques are reviewed to characterize the employee surveillance landscape. Workplace privacy laws, examples of over-surveillance, and the resulting privacy harms are addressed. The review identifies research gaps related to over-surveillance, including the generation of excessive alerts that may obscure meaningful InT indicators. The paper concludes with recommendations to improve workplace surveillance transparency, implement InT training programs to improve organizational detection capabilities, and tune InT tools to detect relevant psychological and behavioral indicators.
Read the full abstract

All authors Haywood Gelman, John D. Hastings, Suvineetha Herath, Quentin Covert

Robotics

Beyond Imitation: Self-Improving Robot Policies via Off-Policy Q-Planning

Varun Giridhar, Anant Khandelwal, Jeremy A. Collins et al. (+2 more)

Behaviour Cloning (BC) has driven remarkable progress in robot manipulation, yet it is fundamentally limited by its inability to self-improve: a policy that fails cannot learn from that failure without additional human demonstrations. Reinforcement Learning fine-tuning offers a path to self-improvement but has proven difficult to scale to the multi-billion-parameter models underpinning modern robot policies. We propose Q-Planning, which equips a large visuomotor BC policy with a small off-policy Q-function. Because a Q-function estimates value rather than imitates actions, it can be trained on the same successful demonstrations as the BC policy and later absorb both successful and failed deployment rollouts, an asymmetry BC does not have. We exploit this asymmetry to enable value-guided action selection at inference (a single-step Q-weighted average over BC draws) and online self-improvement that fine-tunes only the Q-function, leaving the BC weights untouched. On LIBERO and bimanual RoboTwin, ten iterations of self-improvement lift every benchmark score we tested (LIBERO-10 93% to 99%, RoboTwin 83.8% to 91.4%) and shorten successful episodes on the near-ceiling suites (LIBERO-Object, LIBERO-Goal). On two contact-rich bimanual real-robot tasks, the same loop (BC frozen, no human intervention) improves purely from its own deployment rollouts: stack-cups 40% to 90% and insert-wallet 25% to 80% in five iterations, whereas SFT on successful rollouts alone stalls at 55% and 30%. Under an identical online budget Q-Planning is the only method, among Best-of-N, filtered SFT, IBRL, DSRL, and DAWR, that improves stably from failures without training an auxiliary actor.
Read the full abstract

All authors Varun Giridhar, Anant Khandelwal, Jeremy A. Collins, Ignat Georgiev, Animesh Garg

Artificial Intelligence

SENTRY: Deterministic, Intelligent Risk Assessment for IT Change Management

Daniel Arulpragasam, Christer Henrysson, Ella Ly et al. (+2 more)

Technology change management in large financial institutions depends on risk assessments that are accurate, consistent, and auditable. In practice, many institutions still rely on self-reported questionnaires. Those questionnaires are subjective, easy to game, and poor at separating routine changes from the ones that later trigger major incidents. This paper presents SENTRY, a risk assessment platform that replaces questionnaire-based scoring with a deterministic machine learning pipeline built from gradient-boosted decision trees (XGBoost) and hybrid retrieval-augmented generation (RAG). The system combines structured operational metadata, application dependency graphs, and historical incident records with a hybrid semantic and lexical search over historical change requests. The retrieval step captures the risk signal in unstructured change request text, then compresses that signal into a single scalar feature before model inference. That design keeps the model deterministic and preserves per-prediction explainability via SHAP values. Evaluated on enterprise-scale change data, SENTRY achieves a ROC AUC of 0.87 and 85% overall accuracy, and it detects high-risk changes at roughly 3.25 times the rate of the existing process. We close by examining the architectural trade-offs behind this design and what they imply for the use of machine learning in regulated change management.
Read the full abstract

All authors Daniel Arulpragasam, Christer Henrysson, Ella Ly, Deepika Anbalagan, Leo Feng

Game Theory

Half Veto, Half Maximal Lottery, Five-halves Distortion

Qilin Ye

We give a randomized voting rule with expected metric distortion $5/2$, improving the previous best upper bound of $2.753$. Our rule is the equal mixture of a Maximal Lottery [CRWW24] and a lottery obtained by averaging scores in Simultaneous Plurality Veto [KK23] over time. We further show that $5/2$ is optimal within a broader family that allows profile-dependent mixing weights and arbitrary, possibly adaptive, weights over the veto process.
Read the full abstract

Machine Learning

Tydra: An Efficient Hybrid Model for Tabular Data

Mieszko Komisarczyk, Saurabh Mathur, Maurice Kraus et al. (+2 more)

Transformer-based tabular foundation models such as TabPFN achieve strong predictive performance but incur quadratic computational cost with context length. On the other hand, subquadratic SSM-based alternatives such as Hydra trade away accuracy for efficiency. To balance both, we introduce Tydra, a hybrid Transformer-State Space Model (SSM) architecture for tabular in-context learning that interleaves attention and SSM layers. Across 30 OpenML datasets, Tydra reduces inference time by 30% relative to TabPFN while retaining much of its predictive performance. Tydra also outperforms an approximately ten-times-larger Hydra model while providing faster inference. The results indicate that hybrid architectures are a promising direction for tabular foundation models.
Read the full abstract

All authors Mieszko Komisarczyk, Saurabh Mathur, Maurice Kraus, Sriraam Natarajan, Kristian Kersting

Computer Vision

ES-VP : Energy-Shaped Dynamic Visual Prompting for Efficient Model Adaptation

Can Jin, Ying Li, Jingchen Sun et al. (+5 more)

Visual prompting (VP) has emerged as a parameter-efficient method for adapting pre-trained models to downstream tasks. However, existing approaches encounter a trade-off between flexibility and efficiency. Some methods apply a fixed prompt to all images, ignoring individual image characteristics, while others introduce auxiliary networks to generate diverse prompts. Although the latter can improve performance, it also significantly increases parameter usage and the potential for overfitting to specific datasets. Furthermore, the auxiliary networks, combined with inherent biases in pre-trained models, limit scalability and generalization. In this paper, we propose Energy-Shaped Visual Prompting (ES-VP), a novel approach that generates image-specific prompts using low-rank initialization and energy-guided dynamic adaptation, achieving superior performance with fewer parameters compared to single-prompt methods. ES-VP directly utilizes the pre-trained model for adaptive prompt generation, ensuring both parameter efficiency and improved generalization. Extensive experiments conducted on five architectures across fifteen datasets demonstrate that ES-VP consistently outperforms current state-of-the-art (SOTA) single and diverse VP methods. For instance, using the CLIP architecture across four datasets, ES-VP outperforms the SOTA method DAM-VP by an average of 2.6% in accuracy while utilizing 590$\times$ fewer VP parameters, thereby establishing a new benchmark for efficient and generalizable model adaptation.
Read the full abstract

All authors Can Jin, Ying Li, Jingchen Sun, Hongwu Peng, Jiahui Zhao, Yang Zhou, Lei Li, Dimitris N. Metaxas

Computer Vision

Towards Investigating Residual Hearing Loss: Quantification of Fibrosis in a Novel Cochlear OCT Dataset

Julia Dietlmeier, Benjamin Greenberg, Wenxuan He et al. (+10 more)

Objective: Cochlear implants (CIs) are bionic prostheses that restores hearing via electrical stimulation of the auditory nerve. Hybrid CIs, which use electroacoustic stimulation (EAS), combine residual low-frequency acoustic hearing with CI electrical stimulation. Intracochlear fibrosis, which forms in response to the presence of the implant, may impede residual hearing function and gradually reduce the efficacy of EAS. It is therefore a translational objective to study the formation of cochlear fibrosis in rodents, with the goal of reducing fibrotic burden and improving outcomes for CI patients. Methods: We generate and annotate a novel dataset of optical coherence tomography (OCT) images from chronically implanted guinea pigs as part of an ongoing study focused on implant induced fibrosis. Objectively assessing fibrotic burden in this model, with high resolution and repeatability, presents an obvious use case for computer vision methods. Results: We present the results of several state-of-the-art semantic segmentation models and compare their efficacy for identifying cochlear fibrosis and other relevant annotations, using a new library of manually segmented OCT images. Conclusions: We find that the best performance is achieved by using a modified version of the well-known UNET architecture (which we term 2D-OCT-UNET) that operates on the upscaled OCT input resolution. Significance: For the first time, we have successfully applied computer vision techniques to an OCT dataset of implanted cochleae with fibrosis. Using this deep learning model, the cochlear fibrotic burden calculation can be reliably carried out as we verify in our experimental section. The dataset and the project code are available at: https://github.com/juliadietlmeier/CF-OCT-segmentation
Read the full abstract

All authors Julia Dietlmeier, Benjamin Greenberg, Wenxuan He, Teresa Wilson, Rubing Xing, Jordan Hill, Adrienne Fettig, Madeline Otto, Teyhana Rounsavill, Lina A. J. Reiss, Jingang Yi, Noel E. O'Connor, George W.S. Burwood

Artificial Intelligence · Cross-listed from math.OC

Anchored Regularized Direct Least Squares (ARDLS): Integrating Established Prioritization Operators for Priority Elicitation in the Analytic Hierarchy Process

Kevin Kam Fung Yuen

Pairwise reciprocal matrices are fundamental to the Analytic Hierarchy Process (AHP),a decision-making model. While the Direct Least Squares (DLS) method provides an intuitive mechanism for deriving priority vectors without complex transformations, it is susceptible to solution non-uniqueness. Under high levels of inconsistency, such as severe cyclic contradictions, the DLS optimization landscape becomes non-convex, yielding multiple distinct global minima. Consequently, priority rankings become unstable and critically dependent on initial algorithmic guesses. Furthermore, established prioritization operators (POs), including normalization techniques, the Eigenvector method, Singular Value Decomposition, Cosine Maximization, and the Pseudo-Inverse Gram Matrix (the closed-form solution of Weighted Least Squares), frequently generate disparate outcomes. To overcome these structural deficiencies, this paper introduces the Anchored Regularized Direct Least Squares (ARDLS) optimization model as a harmonizing framework. ARDLS integrates uniquely determined established POs as theoretical anchors within a regularization penalty. This integration systematically breaks mathematical symmetries and tilts the optimization landscape to guarantee convergence upon a single, unique global minimum. By minimizing the root mean square variance (RMSV) of the initial baseline vectors, ARDLS effectively unifies these divergent solutions. Comprehensive numerical experiments validate that the framework successfully fine-tune the solution of established POs by reducing RMSV while ensuring strict mathematical uniqueness. The practical utility of the method is further demonstrated through a numerical case study resolving an innovation fund dilemma in FinTech project selection. The proposed ARDLS approach offers a robust alternative to classical AHP across a wide range of decision-making domains.
Read the full abstract

Machine Learning

A Neurosymbolic Approach for Constructing Planning Domain Models from Clinical Narratives

Ranveer Singh, Saurabh Mathur, Michael Skinner et al. (+3 more)

Surgical procedures such as laparoscopic appendectomy are complex, high-stakes processes, yet formalizing their workflows for decision support remains a significant challenge. Inducing probabilistic planning domain models in this setting is particularly difficult due to the lack of structured event data and the prevalence of implicit actions in clinical narratives, which neither empirical symbolic methods nor Large Language Models (LLMs) can adequately address on their own. We introduce NSPIN, a neurosymbolic framework for inducing probabilistic planning domain models from unstructured clinical narratives. Our method extracts and imputes structured event sequences from raw text using a pretrained LLM, then induces a PPDDL model and refines its preconditions with LLM-proposed revisions, guided by empirical validation. We evaluate the approach on 2,660 laparoscopic appendectomy notes written by 9 surgeons. NSPIN yields models that generalize to unseen notes, and expert clinical review indicates its induced knowledge is largely consistent with surgical practice.
Read the full abstract

All authors Ranveer Singh, Saurabh Mathur, Michael Skinner, Prasad Tadepalli, Kristian Kersting, Sriraam Natarajan

Computer Vision · Cross-listed from eess.IV

Toward Vision Language Model-based Assessment of Clinical Quality and Usability of LGE-MR Images for Cardiac Ablation Planning

Bipasha Kundu, Abhishek Chaturvedi, Axel W. E. Wismueller et al. (+2 more)

LGE cardiac MRI is widely used for left atrial fibrosis assessment and ablation planning in atrial fibrillation patients as knowledge of fibrotic tissue regions identified from LGE-MRI is critical for catheter ablation. Often, poor quality images used during ablation planning can cause mis-localization of ablation targets, directly impacting procedure safety and outcome. The decision of whether a scan meets the minimum quality threshold for ablation planning is currently made informally by the reviewing radiologist and is not captured by any automated system, yet it is arguably the most safety-critical output of the image quality assessment (IQA) process. However, variations in image quality caused by noise, motion artifacts, and poor boundary definition significantly compromise the reliability of downstream segmentation and clinical decision-making tasks. Manual quality assessment by expert radiologists is subjective and difficult to scale, while existing automated methods produce scalar scores without interpretable clinical reasoning. In this work, we propose a two-stage vision language model (VLM) framework for clinically grounded image quality assessment of left atrial LGE-MRI. In the first stage, a fine-tuned VLM generates structured radiology-style quality reports predicting five radiologist-defined criteria: Noise, Motion Artifact, LA Boundary Accuracy, PV Region Accuracy, and Under-segmentation Severity. In the second stage, a GPT-based reasoning module maps the predicted quality and reports to a structured quality scores and binary clinical usability decision for ablation planning. We curate a dataset of 60 annotated image slice-text pairs from 20 patients and benchmark four state-of-the-art VLM architectures. InternVL2 achieves the highest criterion-level accuracy (Avg ACC=0.65, PLCC=0.79), while DeepSeek achieves perfect clinical usability agreement (Acc=1.00, kappa=1.00).
Read the full abstract

All authors Bipasha Kundu, Abhishek Chaturvedi, Axel W. E. Wismueller, Richard Simon, Cristian A. Linte

Human-Computer Interaction

From Search Agents to Dissemination Interfaces: Understanding Human Trust in Health Information from Conversational Search

Xin Sun, Rongjun Ma, Xiaochang Zhao et al. (+5 more)

Large Language Models (LLMs) deployed through Conversational User Interfaces (CUIs) are transforming health information-seeking by offering immediate, interactive experiences compared to traditional search engines like Google. However, how trust is influenced by both the types of search agents and the interface used to disseminate the information remains underexplored. This research integrates two mixed-methods studies (lab sessions and interviews) to comprehensively explore trust perceptions in health information across different search agents and dissemination interfaces. In Study 1 (N=21), we investigated trust in health information sourced from ChatGPT and Google across three types of health-related search tasks. Results showed significantly higher trust in health information from ChatGPT, highlighting the promise of LLM-powered conversational search. Building on this, Study 2 (N=20) extended the investigation to explore how the dissemination interface influences trust in LLM-sourced health information by comparing three interfaces: text-based, speech-based, and embodied, all sourcing from the same LLM. Findings revealed significant trust variations across the dissemination interfaces. Interviews from both studies revealed key factors influencing trust in LLM-powered conversational search, including source credibility, participants' search autonomy, and prior knowledge as well as the interaction style and modality. Our findings highlight the potential of LLM-powered conversational search to transform health information-seeking, underscoring the interplay between the credible search agents and the thoughtfully designed dissemination interfaces in shaping trust. These insights are crucial for developing effective, trustworthy LLM-powered health tools to enhance the health information-seeking experience.
Read the full abstract

All authors Xin Sun, Rongjun Ma, Xiaochang Zhao, Janne Lindqvist, Jan de Wit, Zhuying Li, Abdallah El Ali, Jos A. Bosch

Sound

DAMOS: Learning Distortion-Aware Speech Quality Assessment through Explicit Distortion Localization

Naiyuan Li, Li Dong and Diqun Yan

Automatic speech quality assessment aims to predict Mean Opinion Scores (MOS) consistent with human subjective perception and is essential for evaluating speech generation, enhancement, and communication systems. For speech signals, especially synthetic speech, distortions often occur locally, and overall perceptual quality is usually dominated by a small number of perceptually salient distortion regions. However, most existing methods are primarily optimized with utterance-level MOS, which provides only coarse-grained supervision and offer no explicit indication of where perceptually important distortions occur. To address this limitation, we introduce explicit distortion localization as auxiliary knowledge for speech quality assessment. We construct the first partially distorted speech dataset with frame-level distortion annotations and train a localization model to generate distortion cues. Building on these cues, we propose DAMOS, a distortion-aware speech quality assessment framework that integrates localization information into the MOS prediction pipeline. Experiments on multiple public benchmarks demonstrate that DAMOS consistently outperforms existing methods and exhibits strong cross-dataset generalization, validating the effectiveness of explicit distortion localization for speech quality assessment.
Read the full abstract

Robotics

SRL-MPC: Shape-Aware Reinforcement Learned Model Predictive Control

Ruihua Han, Rui Gao, Zhe Liu et al. (+6 more)

Safe and efficient shape-aware navigation in heterogeneous crowds and robot fleets remains challenging. Traditional approaches often assume homogeneous robots, sparse workspaces, simplified geometry, offline computation, or handcrafted parameters to make the problem tractable, which limits their deployment in dense crowd scenarios. Toward this end, we propose Shape-Aware Reinforcement Learned Model Predictive Control (SRL-MPC), a method for safe, efficient, and adaptive navigation in crowds with heterogeneous shapes without geometry simplification. To encode shape-aware safety, we formulate high-order control barrier function (HOCBF) constraints from geometric separation features (GSFs) based on support function transformation. A reinforcement learning (RL) framework then learns a neural policy that reads GSFs and outputs real-time MPC parameter updates, enabling the MPC solver to adapt to neighboring crowd geometries. The key advantage of SRL-MPC is that it preserves the safety structure and generalizability of MPC while integrating the adaptability and intelligence of RL. Experiments in randomized crowd scenarios with arbitrary shaped robot fleets demonstrate the effectiveness, scalability, and robustness of SRL-MPC. The results show that SRL-MPC substantially outperforms representative baselines in safety and adaptability. Project website: https://hanruihua.github.io/srl_mpc_project/
Read the full abstract

All authors Ruihua Han, Rui Gao, Zhe Liu, Xinyi Wang, Chang Chen, Shuai Wang, Qi Hao, Jia Pan, Hengshuang Zhao

Artificial Intelligence

From Attention Masks to Inert Zero-Vector Tokens: OAttention and O-Closure for Token Dynamics

Heyang Gong

Attention masks are relation-level controls: they specify which query–source pairs may interact. They do not provide a representation-carried token state that is non-participating at the attention boundary. We assign each token hidden carrier \(h_i\) an active-presence coefficient \(p_i=\lVert h_i\rVert^2/(\tau+\lVert h_i\rVert^2)\). The same coefficient has two roles: it gates information emitted by token \(i\), and it determines the mass with which token \(i\) enters computations shared with other tokens. OAttention is the support-coupled attention realization of this rule. It gates the receiver output by \(p_i\) and weights source \(j\) by \(p_j\) in both the attention numerator and partition, while retaining the standard score, visibility relation, exponential competition, and value aggregation. This makes the zero-vector token a zero element and yields exact null-receiver, null-source insertion, self-attention insertion, and empty-support properties. The same token-level presence gives local O-components (OFFN, ONorm, and OInject), presence-weighted OStandardize, the O-Closure law \(M(H\oplus0)=M(H)\oplus0\), and an OTransformer by residual and compositional closure. The canonical operator is checked by contract tests and a GPU evaluation. In a zero-fine-tuning retrofit of a cloned pretrained TabPFN v3 regressor, calibrated hidden-carrier OAttention and Full-O variants change mean RMSE by $+0.088\%$ and $+0.177\%$, respectively, over 18 matched dataset–seed cases. A two-block ablation shows that OAttention alone does not preserve a NULL state through ordinary host components, whereas the OTransformer path does. These are scoped tests of exactness, active-path compatibility, and compositional necessity; they do not establish universal no-loss, arbitrary-host closure, learned attraction to the origin, or a general semantics for missing values.
Read the full abstract

Machine Learning

Thermo-FL: Thermal-Aware Robust Federated Fine-Tuning of Large Language Models for Edge AI

Shiva Shrestha, Kazi Shaharair Sharif, Zongxing Xie et al. (+3 more)

Federated fine-tuning enables large language models to adapt on edge devices without centralizing private data, but practical deployments must address hardware instability and adversarial update corruption together. Thermally constrained clients may throttle, slow local training, or delay synchronous aggregation, while Byzantine clients and communication-layer adversaries can corrupt the updates used to form the global model. To address these challenges, we present Thermo-FL, a thermal-aware federated LoRA fine-tuning framework that uses device temperature as an active control signal for local adapter training and sparse update transmission. On the client side, Thermo-FL adjusts the active LoRA-layer fraction and transmitted update density as devices heat or cool, reducing workload under thermal stress. On the server side, Thermo-FL introduces TERRA, a robust aggregation pipeline for dynamically sparse LoRA updates that combines norm filtering, mask-aware directional validation, adaptive active-coordinate clipping, and mask-aware aggregation. We evaluate Thermo-FL using both a large-scale emulator and a Jetson-based physical testbed. In the emulator, Thermo-FL improves robustness under adversarial sparse aggregation and achieves the strongest BoolQ accuracy across clean and attack settings while remaining competitive on GSM8K. In the physical prototype, Thermo-FL stabilizes device temperature, reduces compressed upload size through bitmap sparse encoding, and preserves GSM8K utility under sign-flip/scale and MITM perturbations. These results show that secure edge LLM adaptation should jointly consider hardware behavior, workload regulation, sparse communication, and aggregation robustness.
Read the full abstract

All authors Shiva Shrestha, Kazi Shaharair Sharif, Zongxing Xie, Jiajing Huang, Anhao Xiang, Honghui Xu

Computer Vision

Is Visual Prompting All You Need? Studying VLM Spatial Reasoning under Progressive Visual Scaffolds

Lars Benedikt Kaesberg, Tianyu Yang, Florian Valentin Wunderlich et al. (+4 more)

Vision-language models (VLMs) have advanced rapidly in multimodal reasoning, yet recent work shows that their failures often reflect an interaction between visual grounding and downstream reasoning. What remains less clear is how the visual presentation of a task shapes model performance and failure modes when the underlying reasoning problem is unchanged. We study this question in SPaRC, a benchmark for grid-based visual spatial planning, by introducing lightweight input-side scaffolds that preserve the visual modality while making spatial structure more accessible. Across multiple VLMs, these scaffolds improve task accuracy over the original visual setting by up to 34.0 percentage points and further complement GRPO-based training, yielding up to 4.6 additional accuracy points compared with near-zero gains on the original visual input. Analyses on both end-to-end task solving and object detection show that these gains are closely tied to reductions in grounding-related errors, while rule reasoning remains comparatively challenging. We find that visual presentation is a central factor that determines whether VLM benchmarks measure grounded perception, downstream reasoning, or a mixture of both.
Read the full abstract

All authors Lars Benedikt Kaesberg, Tianyu Yang, Florian Valentin Wunderlich, Terry Ruas, Daniel Kurzawe, Jan Philip Wahle, Bela Gipp

Software Engineering

On the Time and Frequency Domain Representations of Signals for CPS Specification

Claudio Mandrioli, Drishti Yadav and Domenico Bianculli

Specification languages are instrumental to the Verification & Validation of Cyber-Physical Systems (CPSs). Most state-of-the-art specification languages use the time-domain representation of signals, which is not always suitable for describing signal shapes and dynamic behaviours. Instead, fields like control and robotics use the frequency-domain representation to characterise these behaviours. Time-frequency representations combine the capabilities of both domains. We investigate the use of time-frequency representations to specify CPS requirements. We analyse existing taxonomies of CPS requirements to identify which requirement classes can benefit from time-frequency representations. We derive the desiderata for a specification language that uses time-frequency representations and propose Signal-Spectrum Temporal Logic (S2TL), a language enabling assertions over frequency intervals and relations between frequency components. We operationalise the S2TL semantics for monitoring CPS traces, and implement a monitor. We define specification templates for the identified requirement classes and compare time- and time-frequency-domain formulations in terms of applicability, expression fidelity, and noise tolerance of monitoring. We observe that, while time-domain specifications are applicable only to input traces containing step-like changes or using constant interpolation, time-frequency specifications extend their evaluation to traces generated through linear interpolation, and improve tolerance to offset and high-frequency noise, while achieving comparable fidelity to the intended system properties.
Read the full abstract

Human-Computer Interaction

Distilling Black-Box Machine Learning into a Small, Self-Explaining Language Model for Learning Analytics

Chenguang Pan, Airui Meng and Youmi Suk

Learning analytics increasingly relies on flexible machine learning (ML), but the model opacity and the burden of deployment prevent these tools from reaching educational practice. We propose a two-stage fine-tuning pipeline that distills a fitted black-box estimator and its post hoc interpretation (the mentor) into a small, open-weight large language model (LLM; the mentee) that returns an individual-level estimate and explains in natural language. The design is estimator-agnostic and paired with a faithfulness-first evaluation framework that audits every narration against the attribution it claims to describe. We design a simulation study that separates distillation loss from estimator loss by comparing an oracle mentor with a realistic ML mentor. Given an oracle signal, distillation with a two-billion-parameter LLM model is nearly lossless in recovering the effect surface (r > .90), perfectly ranking the important variables, and citing no spurious covariate. Under a realistic estimator, almost all remaining error originates upstream. We find that fluency is no evidence of correctness since narration quality is independent of signal quality, and decision quality collapses toward the majority action in severely imbalanced settings. Applied to a nationally representative dataset, the pipeline recovers the finding that advanced mathematics coursework benefits students least likely to enroll in four-year college the most, with 98.8% of narrations passing the audit and no fabricated quantities. The result is a single fine-tuned LLM that predicts and explains offline on a commodity laptop, so student records never leave the machine.
Read the full abstract

Distributed Computing

GrAND: GPU-based Dynamic Graph Indexes for Approximate Nearest Neighbour Search

Karthik Venkatasubba, Shivendra Deshpande, Shivram S et al. (+1 more)

Modern Approximate Nearest Neighbour Search (ANNS) applications operate over continuously evolving vector collections and require graph indexes that sustain high-throughput searches while incorporating insertions and deletions with high recall. However, most GPU graph indexes are static or provide limited update support. Updates require neighbour discovery, reverse-edge creation, pruning, and deletion-induced graph repair; executing these operations concurrently introduces redundant distance computations and conflicting accesses to shared adjacency lists. Background-rebuild-based deletion further incurs substantial computation, additional memory consumption, and interference with foreground queries. We present GrAND (GPU-based Dynamic Graph Indexes for Approximate Nearest Neighbour Search), a GPU-native collection of dynamic-update algorithms for two popular graph indexes, Vamana and CAGRA. GrAND consolidates graph repair across a batch, eliminating redundant pruning computations, and employs a lock-free find-and-replace strategy for parallel adjacency-list updates. For reliable in-place deletion, GrAND constructs an on-demand reverse graph on the GPU, accurately identifying incoming edges without permanently duplicating the index. We evaluate GrAND on seven real-world datasets across five streaming workloads, comparing it against SVFusion and FreshDiskANN-GPU (our GPU adaptation of FreshDiskANN). GrAND improves overall workload throughput by 2.2x-8.7x and 6.5x-25.4x, respectively, while maintaining high search throughput and recall over sustained updates.
Read the full abstract

All authors Karthik Venkatasubba, Shivendra Deshpande, Shivram S, Jyothi Vedurada

Computer Vision

Human-JEPA: A Human-Centric Vision Model that Perceives and Anticipates

Hui Wei, Licai Sun and Guoying Zhao

Machines that understand humans should perceive the present and anticipate the future. Existing human-centric vision model are pretrained on human images, set the state of the art in static dense perception, so motion and anticipation are out of reach. Here we present Human-JEPA, a human-centric vision model trained on video by anchored forecasting: dense targets are pinned to a frozen copy of the initialization, preventing a silent collapse of dense perception, and block masks are replaced by a pure past-to-future split, avoiding a five-point action tax and a seventeen-point re-identification collapse. Under frozen probes, Human-JEPA leads the pixel-anchored specialists on pose and person re-identification at 2.7 times fewer parameters, conceding high-resolution dense parsing, and its released predictor head is the first that does not degrade anticipation. A single safely adapted model thus serves both halves of understanding humans.
Read the full abstract