Computer Vision
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Tuesday 11 August
Age-related Macular Degeneration (AMD) is the major cause of blindness in the Western world. Its late dry phase is characterised by irreversible atrophic areas, namely Geographic Atrophy (GA). Longitudinal Fundus Autofluorescence (FAF) image acquisitions are currently the main tool for assessing lesion growth over time at the image level. However, due to its highly individualised progression, the evolution of late AMD remains poorly understood. In this work, we propose using Implicit Neural Representations (INRs) to model GA progression at the individual level in a low-data setting. Our approach generates both FAF and GA segmentation at both past and future time points. Among the comparison models, our method achieves competitive segmentation quality across different scenarios, yielding the lowest Mean Absolute Error (MAE) for the GA lesion area and the highest DICE score, without sacrificing FAF image quality. The code is available at https://github.com/SimoneSarrocco/ga-progression-with-inrs.
Computation and Language
Assessing Reliability of BERT-Based Models on Question Answering Tasks
Reliability estimation of large language models is in many cases as crucial as their accuracy, as reliable models are more trustworthy, robust, and suitable for practical applications. Recent advancements in natural language processing (NLP), particularly those based on transformer architectures, have significantly accelerated progress across various NLP tasks. This study focuses on the reliability of transformer-based question answering (QA) models, specifically BERT models and its variants (RoBERTa, ALBERT, DistilBERT). These encoder-only pretrained transformers have demonstrated remarkable accuracy in QA tasks that can be treated as classification tasks. However, their reliability remains underexplored. This study evaluates the reliability of four BERT-based models by assessing response stability under two conditions: (1) internal model variations induced via Monte Carlo Dropout (MCD) and (2) input perturbations through paraphrasing. Using the SQuAD and QuAC datasets, we investigate how dropout rates affect prediction consistency and whether lexical changes impact answer stability. Our findings reveal that RoBERTa maintains higher reliability, whereas AlBERT and DistilBERT exhibit significant inconsistencies. Statistical analyses confirm that enabling MCD during prediction does not disrupt inference dynamics, validating its effectiveness as a reliability metric. These findings underscore the importance of evaluating both accuracy and stability in QA models to ensure stability in real-world applications.
Read the full abstract
Computer Vision
Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation
Wavelet convolution (WTConv) has emerged as an increasingly popular drop-in replacement for standard convolutions, expanding a network's receptive field exponentially with the number of decomposition levels while keeping the parameter count linear. However, its reference implementation is severely memory-bound due to excessive data movement through high-bandwidth memory (HBM). We develop an I/O model of WTConv to characterize this bottleneck and use it to guide three algebraic reformulations: (1) recomputing the inexpensive Haar analysis butterfly on chip, (2) collapsing the multi-level synthesis cascade into a single closed-form pass indexed by output-coordinate bits, and (3) folding learned per-channel scales into the convolution weights. Together, these reformulations enable an I/O-aware fused implementation that substantially reduces HBM traffic. We evaluate the WTConvNeXt configuration across decomposition levels and a broad range of tensor shapes. Despite performing comparable arithmetic, the reference WTConv is substantially slower than the depthwise convolution it replaces. Our reformulation reduces modeled HBM traffic by approximately $2.55\times$, yielding up to a $4.35\times$ training speedup over the reference while roughly halving peak memory usage. Thus, our reformulation preserves the benefits of WTConv while substantially reducing its execution time and memory footprint, removing the systems overhead that previously limited its practical efficiency.
Read the full abstract
Computer Vision
BPG: Balancing Plasticity and Generalization for Domain Incremental Learning
Deep neural networks excel in various tasks but struggle to generalize across evolving data distributions, leading to significant performance degradation under domain shifts. Domain incremental learning (DIL) addresses this challenge by enabling models to continuously adapt while retaining prior knowledge. Among existing DIL approaches, the parameter-isolation paradigm achieves state-of-the-art performance. However, these methods often adopt a one-size-fits-all approach to adapt to new domains, resulting in either insufficient learning capacity or redundant parameters. In this work, we propose BPG, a unified framework that addresses both challenges through two complementary components: BPG-Adapter, which dynamically determines each domain's adapter hidden dimension based on domain-specific feature separability, and BPG-Inference, a soft domain mixture strategy that integrates multiple domain-specific models at test time, mitigating domain ID misselection. Experimental results on DomainNet, CDDB, and CORe50 demonstrate that BPG consistently outperforms uniform adapter-based approaches and hard domain selection strategies, achieving state-of-the-art average accuracy while reducing forgetting to as low as 0.22% on DomainNet.
Read the full abstract
Hardware Architecture
Adaptive Matrix Multiplication for Dynamic Shapes on Ascend NPUs
Matrix Multiplication (MatMul) faces a "generalization crisis" driven by highly dynamic tensor shapes. This crisis is particularly acute on Ascend NPUs, where explicitly controlled architectures and strict physical constraints render existing GPU-centric optimizations ineffective. To resolve this, we propose AdaptCore, an adaptive framework for universally high-performance MatMul on Ascend NPUs. AdaptCore systematically decouples operator optimization into spatial tiling and instruction orchestration. It first maps dynamic shapes into a hardware-aware 2D tiling taxonomy to balance on-chip capacity limits and multi-core parallelism. Furthermore, it integrates a composable optimization library with a deterministic analytical performance model. By mathematically evaluating hardware state mutations, AdaptCore proactively selects and caches optimal implementations, enabling O(1) overhead runtime dispatching. Evaluations demonstrate that AdaptCore delivers a remarkable 1.85x mean speedup across 80,000 input shapes, and achieves up to a 1.48x acceleration in representative end-to-end models over the highly-tuned native vendor library (ACLNN).
Read the full abstract
Computer Vision
Evaluating Semantic and Spatial Guidance for Foundation Model Segmentation of Small-Scale PV in Remote Sensing Imagery
Spatio-temporal PV data are essential for understanding adoption processes in off-grid regions, yet such data remain largely unavailable. Automated segmentation of remote sensing (RS) imagery offers a promising solution; yet, residential PV systems remain challenging targets because of their small size and sparse distribution, resulting in severe target-background imbalance. Vision-language foundation models (FMs) provide a data-efficient paradigm through prompt-based semantic and spatial guidance, but the relative contribution of different prompt types remains unclear. We systematically evaluate SAM3 for small-scale PV segmentation in RS imagery by comparing textual, geometric, and hybrid prompting, under varying supervision levels, training strategies, spatial resolutions, and imaging conditions. Multi-temporal aerial imagery from a large off-grid rural region serves as a study site, with findings validated across three additional datasets. Prompting strategy emerged as the dominant factor governing model behavior. Textual prompting consistently produced the lowest performance and showed the greatest sensitivity to supervision and imaging conditions. In contrast, spatial guidance substantially improved both segmentation accuracy and robustness. Hybrid prompting achieved the highest accuracy and stability, indicating that semantic and spatial guidance provide complementary information. Most performance gains were achieved with only a few hundred annotated samples, demonstrating strong data efficiency. Transfer learning had limited overall impact, with only modest improvements observed for textual prompting under limited supervision. Overall, our findings establish prompting strategy as a key determinant of SAM3 adaptation, robustness, and generalization, highlighting the potential of promptable FMs for scalable PV mapping in data-constrained off-grid regions.
Read the full abstract
Numerical Analysis
High-order $L^2$-Galerkin schemes for the decoupled potential integral equations
We prove Sobolev space well-posedness of the decoupled potential integral equations. Using basic pseudo-differential techniques, discrete stability is obtained for the standard $L^2$-pairing, and we deduce spectral convergence for high-order schemes employing discontinuous basis functions, or with additional element-wise tangential continuity imposed in the vectorial part of the system. The schemes are shown to be much better conditioned than a mature commercial EFIE/CFIE solver. In many cases appearing almost independent of wavenumber, mesh density, and the order.
Read the full abstract
Computer Vision
Beyond Fixed Luminance: Towards Panchromatic and Orthochromatic Image Colorization
Most image colorization systems operate in $Lab$ space by predicting chroma ($ab$) while preserving an input-derived luminance channel ($L$). While effective on standard benchmarks, this fixed-luminance design restricts brightness changes and becomes unreliable when grayscale formation deviates from natural-image luminance, as in historical orthochromatic photography. We propose a luminance-agnostic colorization framework that formulates colorization as full-RGB image editing using a foundation image-editing model. To bridge modern panchromatic and historical orthochromatic conditions, we introduce a mixed grayscale objective that trains the model under both standard luminance grayscale and a red-insensitive grayscale formation. Experiments on COCO, ImageNet, and a multi-instance benchmark show that our method is competitive on standard grayscale inputs and substantially more robust under orthochromatic inputs, with qualitative comparisons and a human study indicating fewer visible color artifacts.
Read the full abstract
Computer Vision
E$^3$mo-Bench: A Scalable Benchmark for Multimodal Evoked and Expressed Emotion Understanding via Bayesian Pairwise Alignment
Understanding both expressed and evoked emotions is critical for multimodal large language models (MLLMs) to achieve comprehensive affect-aware interactions. However, existing benchmarks typically examine expressed and evoked emotions in isolation or are constrained to coarse-grained and incomplete affective characterizations. To bridge this gap, we introduce E$^3$mo-Bench, a scalable benchmark comprising $12{,}314$ question-answer pairs across $2{,}524$ videos with predefined affective perspectives. It evaluates evoked and expressed emotion understanding via $3$ complementary tasks: emotion perception, open-vocabulary recognition, and valence-arousal-dominance (VAD) assessment. To efficiently scale reliable continuous annotations, we propose Bayesian Pairwise Alignment, which aggregates sparse, low-burden pairwise judgments into anchor-referenced VAD estimates. Furthermore, we develop E$^3$mo-Score, a training-free agent that aggregates complementary judgments from a five-model committee to improve VAD estimation. Extensive experiments validate the effectiveness of our framework and expose a pronounced performance skew between evoked and expressed emotion paradigms. These findings, coupled with MLLMs' persistent deficits in fine-grained recognition and dimensional assessment, chart a clear course for advancing multimodal emotional intelligence.
Read the full abstract
Artificial Intelligence
EvoMem: Memory-Augmented Evolution for Code Optimization
Successful mutation strategies in evolutionary code search may contain reusable knowledge that is useful beyond a single run, and in some cases may transfer across related tasks and domains. However, existing LLM-driven evolutionary frameworks largely discard such knowledge, repeatedly rediscovering similar ideas and limiting opportunities for cross-run and cross-task learning. We introduce EvoMem, a persistent memory architecture for LLM-based evolutionary program search that captures and reuses candidate mutation knowledge. EvoMem converts successful mutation events into structured, task-aware advice for future runs. It operates in two phases: after each run, it extracts and stores promising ideas with provenance, and during subsequent evolution, it retrieves a small set of relevant instructions based on the current task and program context to guide mutation. Across geometric optimization, multi-hop question answering, GPU kernel optimization, and related benchmarks, our experiments show positive average improvements in target metrics or search speed for most evaluated settings, while also revealing variability across tasks. Overall, EvoMem provides evidence that persistent memory can reduce some redundant exploration and improve the reuse and adaptation of successful strategies in LLM-driven evolutionary search.
Read the full abstract
Social and Information Networks
Shaping the notion of #wellbeing in the therapy culture context: an analysis through Instagram narratives
The aim of this paper is to (1) identify textual and visual themes and sub-themes associated with the #wellbeing hashtag on Instagram, (2) assess their varying levels of engagement and (3) investigate gender bias present in the analysed visual narratives. This study employs a range of big data analysis techniques to investigate various dimensions of wellbeing on Instagram. Initially, a sample of 9,844 posts was processed using data mining and text analysis methods to identify and categorise significant hashtags, which facilitated the classification of posts into thematic clusters through hierarchical clustering. Engagement was then assessed using non-parametric statistical tests. Additionally, computer vision models were used to analyse and classify visual narratives, grouping images into communities based on visual similarities. Finally, gender representation in the images was examined using object detection models. The study reveals that the discourse around #wellbeing on Instagram is predominantly feminised and focuses primarily on mental, psychological and spiritual aspects. Therapeutic and positive psychology narratives are the most prevalent and engaging, while physical activity and nutrition play a relatively secondary role. Two major macro-narratives emerge: firstly, mental health and emotional wellbeing, often featuring hashtags related to therapy, self-discovery, spirituality and motivational quotes; secondly, though less prominently, themes and visual narratives concerning physical activity and healthy habits, emphasising exercise and nutrition.
Read the full abstract
Artificial Intelligence
ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation
Autonomous chemistry increasingly depends on environments in which agents can repeatedly act, observe, and adapt.Physical laboratories provide essential real-material evidence but are costly to repeat and difficult to use for tightly matched interventions, whereas most digital environments keep the underlying experimental world largely fixed. We introduce ChemWorld, a programmable chemical environment in which reusable process and observation components are compiled into executable worlds. ChemWorld separates the public experimental contract available to an agent from evaluator-owned chemical and material laws. Researchers can therefore vary world composition and operating conditions, or change a single hidden law while holding the public task and interaction conditions fixed. Transactional execution records operations, failures, resource changes, and state transitions, allowing complete environment-action trajectories to be replayed exactly and audited. Full-census qualification covered the reference registry, 52 generated compositions, and module, interface, compilation, and invalid-action tests. Eight deterministic experimental cases demonstrated shared lifecycle semantics, failure recovery, and exact replay, while six parent-child world-fork pairs isolated the effects of single private-law interventions under matched public conditions. An independent agent also completed a full lifecycle in a non-reference world through the same public interface. Within the declared component and model domain, ChemWorld provides a controlled and replayable substrate for studying experimentation across systematically varied chemical worlds, complementary to physical-laboratory evidence and calibration.
Read the full abstract
Robotics
Dual Stress: Runtime Safety Monitoring for Safety-Constrained MPC Navigation
Runtime hazard monitors for autonomous naviga- tion are conventionally built from geometric quantities: predicted clearance, time to collision, and required deceleration. A model-predictive controller that enforces safety through explicit con- straints computes, as a by-product of every control step, a second information channel that such monitors ignore: the Karush-Kuhn-Tucker multipliers of its constrained optimization, which measure the marginal control effort spent to maintain safety against each obstacle. This paper evaluates whether a horizon-weighted sum of those multipliers, a dual stress signal, provides a hazard monitor complementary to the geometric warnings the same state already supports. We compare it against a battery of fifteen geometric detectors tuned to a matched false-alarm budget, on preregistered held-out crossing scenarios driven through a physics simulator. The stress alarm actionably flags 4.7 times as many collisions missed by the entire geometric battery as the geometric battery flags in return (85 versus 18); combined, the two channels warn of three quarters of the collisions for which braking remained feasible, against under half for the geometric battery alone.
Read the full abstract
Computer Vision
MVTrack: Ultrafast Appearance-Free Moving Object Tracking from Compressed Bitstreams
Deploying modern video trackers at scale is bottlenecked by the computational cost of RGB-based object detectors. To this end, we present MVTrack, an ultrafast tracker for moving objects that operates directly on H.264 bitstreams. MVTrack combines MVDet, a lightweight detector for motion vector fields, with MVLink, a minimalist kinematic association module. On VIRAT, MVTrack outperforms YOLO26n while using 60$\times$ fewer parameters, requiring 40$\times$ fewer FLOPs, and reducing CPU latency by 8.6$\times$. These results demonstrate that compressed video data alone can enable accurate and scalable surveillance tracking, thereby bypassing the need for pixel reconstruction.
Read the full abstract
Software Engineering
Extending Triple Graph Grammars to Formalize Complex View Definitions on Families of Models - Long Version
View-based development is an important technique to manage complexity and facilitate cooperation in the development of modern and complex systems of systems. This approach places high demands on view definition languages used in the development process. First, such a language needs to be approachable and extremely versatile so that developers from diverse backgrounds are able to use it and can define views for a diverse set of different tasks. Yet, the language also needs to have a precise and formal semantics so that it is possible to integrate the work performed on various views in a controlled manner to obtain a coherent overall system. In this paper, we continue previous work of equipping the NeoJoin view definition language with a formal semantics that is based on triple graph grammars (TGGs). This paves the way for obtaining automated and incremental synchronization procedures between models and views that come with high formal guarantees for their behavior. Simultaneously, our work serves as a further case study of the expressivity and usability of TGGs. We identify one gap, namely convenient support for the translation of overlapping queries from the view definition language, and tackle that gap by introducing a skip semantics for TGG rules, i.e., allowing to skip certain actions a rule prescribes, depending on the application context.
Read the full abstract
Networking
Multi-UAV Tracking Evaluation Using 5G Uplink Signals on an O-RAN ISAC Simulation Testbed
We evaluate multi-target detection, association and tracking end to end on an O-RAN simulation testbed built from OpenAirInterface, FlexRIC and Sionna RT that repurposes the 5G NR uplink sounding reference signal as a passive radar waveform, and against what a counter-UAS command-and- control (C2) consumer requires rather than by detection alone. Three UAVs differing in altitude, velocity and radar cross section (-8 to -20 dBsm) fly one bistatic pair with an 8-element planar receive array. Once every target is detected the binding limit is contention, not sensitivity - two targets share one nearest detection in about 74% of coherent processing intervals, and targets are detected far more often than they are tracked. Elevation from that array cannot separate targets sharing a range-Doppler cell, but it decides association in 58% of intervals. Concurrent tracks are exported from the RAN Intelligent Controller (RIC) xApp to a C2 fusion node over a SAPIENT interface carrying a calibrated detection confidence, validated at schema level against a mock fusion node. The evaluation is emulation-only on one geometry with 10 noise seeds, the mechanisms are characterized and analyzed to identify the key factors influencing multi-target tracking performance.
Read the full abstract
Robotics
StageWAM: Joint-Embedding Stage Prediction for World-Action Models in Robot Manipulation
Generalist robot policies aim to map multimodal observations and linguistic task instructions to actions across diverse tasks. However, existing methods typically represent the future as a fixed, short video-action chunk. This short-term future captures local scene evolution for action execution, but it does not explicitly describe the stage-level future that specifies how a task should progress from its current stage to the next. We therefore distinguish two complementary futures for robot manipulation: a short-term physical future to capture local scene evolution and a stage-level semantic future to represent task progress. We introduce StageWAM, which augments a Motus-based World Action Model (WAM) with Stage-JEPA, a goal-conditioned Joint-Embedding Predictive Architecture (JEPA) predictor. Given the current observation and task instruction, Stage-JEPA uses a frozen V-JEPA2 encoder to extract the current-state representation and predicts the latent target of the next inferred stage. Across 50 RoboTwin 2.0 tasks in clean and randomized environments, StageWAM achieves 90.25% overall success and reduces the mean number of execution steps in successful rollouts by 5.97% relative to the strongest baseline.
Read the full abstract
Computers and Society
Technology, education and critical media literacy: potential, challenges, and opportunities
This study examines the impact of technology within media education, media literacy, and educommunication, and explores how these fields are perceived and understood by students and academic experts, focusing on the development of critical competencies and critical media literacy. Based on semi-structured in-depth interviews with leading experts in the field of critical media literacy, and a survey conducted with 141 university students in Communication and Education programs, this study explores how recent technological advances are linked to challenges in information consumption-such as disinformation, fake news, incidental exposure to information, and deepfakes-as well as the challenges and opportunities these issues present within educational contexts. The results reveal that, although such technologies provide opportunities to improve teaching-learning processes, their inclusion in the curriculum is limited and often superficial. In addition, we identify shortcomings in how teachers are trained to manage these tools effectively, hindering the development of critical thinking by students. The conclusions suggest a need for critical media literacy that not only educates students in the use of technologies but also allows them to question and evaluate the content they consume, all within an ethical and reflective framework that promotes participatory and critical citizenship.
Read the full abstract
Machine Learning
Path Integral Value Matching for Linear Quadratic Stochastic Optimal Control
Linear Quadratic Stochastic Optimal Control (LQ-SOC) establishes a fundamental framework for steering noisy dynamical systems and has recently gained renewed interest in the machine learning community. However, current state-of-the-art policy-based methods suffer from prohibitive computational costs and instability due to their heavy reliance on full-trajectory simulation. To overcome these limitations, we propose a paradigm shift toward a value-based approach by revisiting Path Integral Control (PIC). Although standard PIC suffers from the same high-variance bottleneck as policy-based methods, we discover that by truncating and marginalizing the original path integral formulation, we can derive a temporal recursive form of the value function. Building upon this theoretical foundation, we propose the Path Integral Value Matching (PI-VM) algorithm. Specifically, we employ temporal-difference learning to approximate the recursive value dynamics, and further integrate the Girsanov theorem with experience replay to enable off-policy training. We benchmark PI-VM against SOTA policy-based methods across various SOC benchmarks and sampling tasks. Empirical results demonstrate that PI-VM matches SOTA precision with an order-of-magnitude efficiency gain in low-dimensional settings, while effectively mitigating mode collapse in high-dimensional scenarios. Consequently, PI-VM offers a scalable solution for solving complex SOC problems.
Read the full abstract
Artificial Intelligence
SkillLens: Visual Skill Cards for Retrieval-Augmented GUI Action Prediction and On-Policy Distillation
Graphical user interface (GUI) agents can reuse past interaction experience. We study how illustrated procedures support action prediction and students deployed without retrieval. We introduce SkillLens, which represents procedures, applicability cues, and interface views in a shared Visual Skill Card (VSC) schema. Retrieved cards provide a frozen executor with reference evidence alongside the current screen. CardDistill uses this schema to construct privileged teacher context from labeled training steps for on-policy distillation; the student learns from native inputs and requires no retrieval at deployment. With matched training data and update counts, students trained with full-card teachers improve step success rate by an average of 5.04 percentage points over those trained with compact-text teachers across three training seeds on 119 tasks from ten websites absent from training. An independent runtime control uses training-source memory and pre-action retrieval queries: full cards achieve 12/64 correct actions (18.75%) versus 4/64 (6.25%) for text-only cards. These results establish the value of the complete VSC teacher-context package for retrieval-free students and provide evidence for pre-action retrieval of illustrated procedures.
Read the full abstract
Social and Information Networks
A Graph Approach to the Academic Publishing Network: A Heterogeneous Model and Structural Screening over OpenAlex Open Data
The academic publishing ecosystem is a vast, heterogeneous network of works, authors, institutions, journals, and topics. Traditional scientometrics reduces it to isolated tabular indicators (h-index, Impact Factor) that ignore topological context and are not designed to capture coordinated illegitimate practices. Building on our companion review, which proposed graph analysis of publishing integrity, this paper implements that approach. We define a heterogeneous multivariate graph model over OpenAlex open data (seven node types, seven edge types) and a methodology based on projections (citation and co-authorship networks), interpretable structural metrics, community detection, and three screening detectors of anomalous publishing patterns. We deliberately avoid binary classification: detectors return ranked candidates with explicit structural evidence for human assessment. On the institutional corpus of VSB - Technical University of Ostrava (2020-2025) with its one-hop citation neighbourhood, community detection recovers real research groups, centralities identify cross-disciplinary bridges, and the screenings flag dense co-authorship cliques, locally closed citation loops, and thematically isolated venues. On a second, venue-centric corpus with external ground truth (journals delisted by Scopus and DOAJ) and size-matched controls, a naive case-control design yields seemingly strong but spurious detectors (a prominence confound), whereas after matching the only robust signal is the breadth of disciplinary scope (AUC 0.70); an open graph-based prestige measure (PageRank over the journal citation network) tracks a JIF proxy while being an order of magnitude more resistant to citation gaming than count-based indicators. We release the method as the open-source library apnet with a reproducible CLI workflow and a web interface; the analysis runs on commodity hardware in minutes.
Read the full abstract
Computers and Society
The GenAI Catch-22: Use of Generative Artificial Intelligence in Norwegian Newsrooms During the 2025 Parliamentary Election
The increasing use of Generative Artificial Intelligence (GenAI) in journalism raises concerns about possible detrimental effects both on journalism and its democratic function. We explore these risks through a case study of GenAI in Norwegian Newsrooms during the 2025 parliamentary election campaign. Based on interviews with managers and journalists over a ten-month period, we analyse how ambitious visions fared in the face of technological and practical challenges. We highlight the risk of an internal threat stemming from the journalists' own use of AI, contrasting the dominant focus on external disinformation threats. We show how newsroom managers shared sociotechnical imaginaries resulting in unrealistically optimistic beliefs about the capabilities of the technology and the pace of development, leading to plans for audience-facing GenAI services collapsing and giving way to more mundane uses of GenAI tools internally in the newsrooms. Furthermore, we identify a vulnerability in the newsroom's resilience against GenAI influence: a GenAI Catch-22. In order to monitor the GenAI tools and prevent errors and undue influence, newsrooms rely on human expertise. But by using GenAI extensively, the newsrooms risk a deterioration of human expertise, preventing them from monitoring the GenAI systems adequately.
Read the full abstract
Systems and Control
Variable-Step Time-Delay Control for Proactive Aperiodic Spacecraft Attitude Control
This paper addresses proactive aperiodic spacecraft attitude control under model uncertainty, environmental disturbances, and actuator degradation. To this end, we develop Variable step Time Delay Control (VTDC), which jointly designs robust feedback control and control-update scheduling rather than treating them as separate components. Inspired by adaptive stepsize regulation in Runge Kutta integration, VTDC structures the local control error to scale with the realized timestep. Using sliding manifold based Time Delay Control, the resulting System Time Delay Error (TDE), which reflects local model and uncertainty variations, is shown to be quadratically bounded by the control interval. This relation yields a closed form feedback law that enlarges or reduces the subsequent interval to regulate the TDE magnitude. The next update time is therefore determined algebraically without continuous trigger monitoring, future state prediction, or iterative search. The resulting variable step closed loop admits bounded timesteps and step ratios, excludes Zeno behavior, and renders the sliding variable uniformly ultimately bounded. Nonlinear spacecraft attitude control simulations demonstrate accurate tracking with low scheduling cost under representative uncertainties and disturbances.
Read the full abstract
Artificial Intelligence
Rule of Thumb: Explaining Artificial Intelligence Systems using Partial Information
Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ”Rule of Thumb” (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI in: (a) zero-shot classification using large language models (LLMs), (b) auditing of opaque AI systems without model access, and (c) the use of AI in scientific discovery. Additionally, RoT meets specific requirements from leading AI regulations, provides a familiar interface and visualisations for XAI practitioners, is model-agnostic, and is substantially faster than alternatives. Code available at: https://github.com/KaiRawal/Rule-of-Thumb-Explaining-Artificial-Intelligence-Systems-using-Partial-Information
Read the full abstract
Artificial Intelligence
Compositional Benchmark Synthesis for Hierarchical Human Action Recognition
Recognizing human behavior across levels of abstraction, from atomic actions to long-horizon intentions, requires data annotated along a semantic hierarchy. Large corpora provide isolated, atomically labeled clips without temporal composition, whereas recorded composite-activity corpora offer shallow, domain-narrow, fixedhierarchies. A benchmark-generation and evaluation frameworkis proposed that synthesizes a four-level hierarchical-intention benchmark, spanning actions, activities, low-level intentions (LLIs), and high-level intentions (HLIs), from a flat single-label action corpus while retaining real pre-extracted features at the action level. Episodes are assembled by a transition model under a subject-consistency constraint, and a coverage-aware sampler reduces the subject usage Gini from 0.566 to 0.248. Synthesizing such a benchmark raises a circular-supervision risk that recorded datasets avoid: if the rules generating the episodes also govern the evaluation, models can succeed by recovering the generator rather than through genuine reasoning. Validity is addressed by design, holding sequence-generation rules disjoint from the first-order-logic rules used at evaluation. The instantiation yields 15,002 episodes. Four reference baselines from different model families characterize difficulty, not as recognition methods. A compositional held-out gap of 0.13 to 0.17 macro-F1 appears across all baselines, including a graph-aware model that recognizes best yet does not close the gap, indicating a structural property of the benchmark rather than a model artifact. A logic-free baseline still violates the held-out semantic rules above their intrinsic data rate, and the order-destroying control changes macro-F1 within seed variation, serving as a generator-consistency check. Theontology, transition model, and generator are released so the benchmark can beregenerated and extended.
Read the full abstract
Computer Vision
FADE: From Passive Verification to Active Discovery in Counterfactual Video Understanding
Counterfactual video understanding evaluates whether models grasp physical and commonsense regularities. However, existing multiple-choice question (MCQ) benchmarks inadvertently leak target events through their questions and candidate options. This reduces the core challenge from active discovery to text-guided verification. In this paper, we present FADE, an effective training framework for counterfactual discovery and explanation. Our method is built on an evidence-first, two-stage training paradigm. First, evidence-internalized supervised fine-tuning grounds the model's predictions in decisive visual anomalies. Second, we apply a fading-anchor reinforcement learning strategy that progressively removes textual guidance, compelling the model to independently discover and explain evidence. To rigorously evaluate this capability, we also introduce an effective pipeline that converts existing MCQ datasets into aligned MCQ, open-ended question answering (OQA), and captioning tasks without requiring additional data curation. Our simple approach yields strong results. Using Qwen3-VL-8B as the baseline, FADE achieves state-of-the-art strict paired scores across all three tasks on DualityVidQA-test and IPV-Bench, outperforming GPT-5.6. In specific, when transitioning from constrained MCQs to unconstrained OQA and captioning, our model demonstrates remarkable robustness. Its performance retention is 90.4% and 67.4% on DualityVidQA-test-substantially higher than the 48.1% and 30.7% retained by GPT-5.6. We hope this simple framework can serve as a solid baseline for future research in unconstrained counterfactual video understanding.
Read the full abstract
Human-Computer Interaction
Playable Pressure: Affective Dramaturgy and Selective Realism in the Design of a VR Emergency-Response Serious Game
Professional simulations stage not only procedures but models of what should command attention, which emotions belong in competent practice, and whose distress becomes part of the task. This article develops affective dramaturgy through critical design-document analysis of a virtual-reality emergency-response project. The corpus comprises two non-public production records. We identify six families of specified pressure and examine how sensory staging, proximity, trigger authority, task conflict, and response allocation imply a selectively receptive triage professional. The documented design can make emergency work morally and socially crowded, yet it can also turn grief, vulnerability, and mental-health-coded behavior into adjustable difficulty. We propose answerability at two levels-in-play response and post-play debriefability-and derive case-based questions about occupational purpose, representation, adaptation, and accountability. These questions are sensitizing propositions, not a validated framework for user effects or emergency practice.
Read the full abstract
Cryptography and Security
A Gateway Architecture for Enterprise MCP Authentication: Unifying Heterogeneous Auth, Identity Delegation, and the User / Non-User Persona Problem
The Model Context Protocol (MCP) has become the de-facto interface for connecting LLM agents to enterprise tools, and adoption has been explosive: within a year, large organizations went from zero to dozens of internally built MCP servers. That speed created a governance crisis. Each team implemented authentication independently – some with no auth, some with API keys, some with full OAuth – producing a fragmented landscape with no consistent way to authorize callers, track who did what, or offboard a departing employee across the fleet. This paper reports an industry deployment that resolves the crisis with a centralized MCP gateway: a single aggregation, governance, and authentication layer that fronts every downstream MCP server. We make four contributions grounded in production experience. First, a two-axis authentication model crossing persona (interactive user vs. automated non-user) with credential type (no-auth, static/dynamic API key, PKCE, client credentials, platform app-context). Second, a gateway authentication layer supporting three enterprise SSO grants and three token-provisioning models: Bring-Your-Own-Token, Generate-Your-Own-Token, and delegated OAuth via RFC 8693 token exchange. Third, three end-to-end identity flows – User-to-OAuth2, Non-user-to-Service-Account, and User-to-Service-Account – composing client, gateway, and server. Fourth, the deployment evolution from CDN/WAF/edge perimeter to private MCP tunnels and enterprise-wide connectors. The architecture is in production, fronting dozens of MCP servers across web, desktop, custom-SDK, and low-code clients.
Read the full abstract
Computer Vision
Where To Look? : Causal Tracing of Vision Encoders in VLM
Vision-language models can describe an image with remarkable accuracy, yet a more fundamental question remains unanswered: what visual information actually drives their answers? In this work, we investigate this question through causal tracing, and we observe that highly causal vision tokens often lie outside the target region. Extending the analysis to larger vision-language models reveals a similar pattern across models and corruption settings, suggesting that strong multimodal performance does not necessarily imply spatially localized causal representations. We further investigate: can these models preserve visual structure when appearance cues are removed? and find that visual cues are exploited to understand visual structures. Together, our experiments expose a gap between seeing, using, and reasoning over visual structure, and provide a causal framework for studying how visual information is transformed, preserved, and ultimately used by modern vision-language models.
Read the full abstract
Numerical Analysis
Successive Schur-Riesz Analysis for Approximation
Many approximation methods enlarge a trial space by adjoining function blocks generated by different operators. Exact redundancy and strong cross-level interaction can make coefficients nonunique and render pairwise or diagonal-dominance tests needlessly pessimistic. For \(V_m=\sum_{\ell\leq m}S_\ell(E_\ell)\) in a Hilbert space $\mathcal H$, we quotient coefficients representing the same function and control successive orthogonal innovations to obtain Riesz bounds independent of \(m\). The setting includes factored operators \(S_\ell=T_\ell\circ\cdots\circ T_1:E_\ell\to\mathcal H\), with compatible intermediate spaces, but the theorem allows arbitrary bounded \(S_\ell\). The same constants control approximation, truncation, perturbation, and levelwise error. A block Schur complement identifies the intrinsic new dimension and gives the exact reduction in squared best-approximation error, leading to a constructive enrichment procedure. Nonstationary and lifted examples give positive intrinsic bounds where diagonal-dominance estimates are negative or labelled Gram matrices are singular; adaptive and recycled-subspace calculations illustrate the distinct roles of representation stability and application-specific utility.