# Research Papers — 2026-04 · bundle page 11 of 233 > Full-text markdown of papers #501–550 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/11 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/10.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/12.md) ### Interpretable rainfall modelling reveals rapid reorganisation of Amazonian rainfall under vegetation loss Authors: Lilly Horvath-Makkos, Fayyaz Minhas Published: 2026-04-29 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2605.10948v1 Understanding how vegetation loss alters rainfall remains a major challenge in climate and hydrological science, as deforestation modifies precipitation through heterogeneous, seasonal and nonlinear land-atmosphere feedbacks. Existing models struggle to capture these dynamics: convection is parameterised at coarse scales, tipping behaviour is poorly constrained, and rainfall-deforestation analyses are limited to multi-decadal timescales. Therefore, many approaches resolve correlations rather than causal effects, limiting our ability to anticipate hydrological disruption. Using a neural-network model for hourly rainfall prediction, combined with pathway diagnostics and sensitivity analyses, we examine how vegetation perturbations reorganise rainfall across space, intensity regimes, and timescales under deforestation. We assess whether the model captures physically consistent dependencies linking vegetation, atmospheric state, and precipitation, and whether sustained canopy loss induces threshold behaviour. The model accurately predicts rainfall occurrence and intensity (Spearman = 0.84, F1 = 0.93, ROC-AUC = 0.98) and learns temporally ordered dependencies aligned with ecohydrological theory. Sensitivity analyses reveal rapid, asymmetric responses to vegetation loss: heavy rainfall (20-50 mm/h) declines by up to 7% under sustained deforestation, while light rainfall (0.1-1 mm/h) increases by 4%. Rainfall entropy rises by 1.3%, and dry-season intensity increases by 0.3-0.5% per 0.5% forest-cover loss, with strongest impacts in the north-western Amazon and Andean foothills. Threshold analysis reveals a sharp decline in precipitating area fraction after 2-3 months of sustained vegetation change in sensitive regions. These results demonstrate that data-driven approaches uncover process-relevant land-atmosphere coupling and highlight growing hydrological vulnerability in the Amazon. --- ### Interpretable EEG Microstate Discovery via Variational Deep Embedding: A Systematic Architecture Search with Multi-Quadrant Evaluation Authors: Saheed Faremi, Andrea Visentin, Luca Longo Published: 2026-04-29 Categories: cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2605.10947v2 EEG microstate analysis segments continuous brain electrical activity into brief, quasi-stable topographic configurations that reflect discrete functional brain states. Conventional approaches such as Modified K-Means operate directly in electrode space with hard assignment, offering no learned latent representation, no generative decoder, and no mechanism to decode latent configurations into verifiable scalp topographies, limiting both model transparency and interpretability. To address this, we present a Convolutional Variational Deep Embedding (Conv-VaDE) model that jointly learns topographic reconstruction and probabilistic soft clustering in a shared latent space. Conv-VaDE enables generative decoding of cluster prototypes into verifiable scalp topographies, replacing opaque hard partitioning with probabilistic soft assignment. A polarity invariance scheme and a four-dimensional grid search over cluster count (K from 3 to 20), latent dimensionality, network depth, and channel width are conducted to systematically reveal how each architectural design choice shapes the quality, stability, and interpretability of learned EEG microstate representations. The model is evaluated on the LEMON resting-state eyes-closed EEG dataset with ten participants using topographic template formation, clustering stability, and global explained variance (GEV). The architecture search reveals that depth L = 4 appears consistently across all 18 best-performing configurations, yielding a best-case GEV of 0.730 and a silhouette of 0.229 at K = 4 across the model sweeps, where moderately deep networks with compact channel widths and small latent dimensionality dominate across the full K range. These results establish that principled architecture search, rather than model scale, is the key to interpretable and stable EEG microstate discovery via variational deep embedding. --- ### Scaling Mobile Agent Systems: From Capability Density to Collective Intelligence Authors: Bowei He Published: 2026-04-29 Categories: cs.DC, cs.CL, cs.MA, cs.NI arXiv: https://arxiv.org/abs/2605.08124v1 Mobile agent systems are emerging as a key paradigm for enabling intelligent applications on edge devices and in AIoT ecosystems. However, their scalability is fundamentally constrained by limited on-device computation and fragmented intelligence across devices. In this work, we propose a unified research agenda for scaling mobile agent systems along two complementary dimensions: (1) improving capability density of individual agents through compact foundation model design and compression, and (2) enabling collective intelligence via communication-rich multi-agent collaboration. Building on recent model and infrastructure advances, this vision aims to transform isolated mobile agents into a distributed intelligent system that is efficient and scalable. --- ### SAMamba3D: adapting Segment Anything for generalizable 3D segmentation of multiphase pore-scale images Authors: Rui Zhang, Xianzhi Song, Linqi Zhu, Branko Bijeljic, Gensheng Li, Martin J. Blunt Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00916v1 Reliable segmentation of multiphase pore-scale X-ray images of rocks is necessary to quantify fluid saturation, connectivity, and interfacial geometry. However, current 3D segmentation methods are typically dataset-specific, requiring retraining or extensive fine-tuning whenever rock type, fluid pattern, scanner, or acquisition conditions change. Foundation models such as the Segment Anything Model (SAM) provide strong 2D boundary priors, but they are not directly applicable to 3D data. We present SAMamba3D, a parameter-efficient framework that adapts a largely frozen SAM encoder to generalizable 3D pore-scale segmentation by coupling it with Mamba-based volumetric context modeling and progressive cross-scale feature interaction. For sandstone and carbonate datasets, with different fluids, wettability, and scanning conditions, SAMamba3D matches or outperforms current 3D baselines while reducing the need for case-specific retraining. The resulting segmented images preserve physically meaningful descriptors, including fluid saturation, connectivity, and interface morphology, enabling more reliable and rapid analysis of large 3D multiphase images. --- ### Learning Dynamic Evidence Routes for Vision Transformer Probing Authors: Zice Wang, Zhenyu Zhang Published: 2026-04-29 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00915v2 Probing frozen vision transformers typically uses permutation-invariant aggregation (GAP or $\texttt{(CLS)}$), treating patch tokens as an unstructured set. Content-dependent probes such as self-attention are useful accuracy controls, but they do not expose a fixed token schedule or fixed position weights for auditing. We introduce $\textbf{SSMProbe}$, an explicitly inspectable probe that replaces invariant pooling with a Sinkhorn-learned evidence route followed by a diagonal S4 decoder. The S4 decoder is a linear time-invariant (LTI) system whose final state has fixed, position-dependent coefficients, so the probe-induced routed sequence can be audited as a concrete object rather than inferred only from accuracy. Our central measurement is the geometry of routed evidence: which patch tokens are moved to influential positions by this diagnostic, whether those tokens form spatially organized regions or random-like dispersed sets, and how the fixed S4 kernel weights them. Across MAE, BEiT, DINOv2, and supervised ViT, this route geometry separates MAE's dispersed, nearly random-like routes from the more spatially organized routes of BEiT, ViT, and DINOv2, with DINOv2 retaining a distinct strong $\texttt{(CLS)}$ profile. SSMProbe uses the mathematical transparency of state-space models to turn a frozen ViT readout into an auditable evidence-routing analysis. --- ### Rethink MAE with Linear Time-Invariant Dynamics Authors: Zice Wang Published: 2026-04-29 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00915v1 Standard representation probing for visual models relies on mathematically permutation-invariant operations like Global Average Pooling (GAP) or CLS tokens, treating patch representations as an unstructured bag-of-words. We challenge this paradigm by demonstrating that token order is a critical, exploitable dimension in frozen visual representations (e.g., MAE, BEiT, DINOv2, and ViT as CLS-ablation extreme). We propose SSMProbe, a probing framework driven by a State Space Model (SSM). Operating as discrete Linear Time-Invariant (LTI) dynamical systems, SSMs act as permutation-sensitive probes where sequence order strictly dictates the final state due to inherent memory decay. Formulating token ordering as an information scheduling problem, we compare fixed scan heuristics against a differentiable soft permutation (Sinkhorn-based) learned from downstream supervision. Evaluations on standard and fine-grained classification benchmarks reveal a striking order gap: while fixed scans fail dramatically on highly localized patch features, our learned soft permutation successfully extracts highly competitive performance from otherwise heavily localized patch sequences. We find that pre-training objectives fundamentally shape token structure: DINOv2 concentrates global semantics in optimized CLS tokens leaving patches hyperspecialized, pure MAE preserves distributed representations with heterogeneous patch informativeness, and ViT represents a supervised CLS-dominated extreme. BEiT occupies middle ground. This heterogeneity is order-dependent -- meaning the SSM probe's performance depends critically on which tokens are placed at which temporal positions -- and is not merely a topological property of the spatial grid. SSMProbe's learned routing effectively discovers and exploits this heterogeneity, offering a powerful new diagnostic lens for visual representation analysis. --- ### The Cost of Consensus: Isolated Self-Correction Prevails Over Unguided Homogeneous Multi-Agent Debate Authors: Blaž Bertalanič, Carolina Fortuna Published: 2026-04-29 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2605.00914v1 Multi-agent debate, where teams of LLMs iteratively exchange rationales and vote on answers, is widely deployed under the assumption that peer review filters hallucinations. Yet the failure dynamics of homogeneous debate remain poorly understood, therefore we report findings from a controlled empirical study of teams of $N{=}10$ homogeneous agents (Qwen2.5-7B, Llama-3.1-8B, Ministral-3-8B) across $R{=}3$ debate rounds on two high-difficulty benchmarks (GSM-Hard and MMLU-Hard). We compare peer debate against isolated self-correction and a stochastic noise control that injects rationales from unrelated problems. We decompose debate failure into three model-dependent pathways: sycophantic conformity, where agents uncritically adopt majority answers (modal adoption up to 85.5%); contextual fragility, where peer rationales destabilize previously correct reasoning (vulnerability rate up to 70.0%); and consensus collapse, where plurality voting discards correct answers already present in the generation pool (oracle gap up to 32.3 percentage points). Ablations over communication density ($K \in \{2,4,9\}$) and sampling temperature ($T \in \{0.4, 0.7\}$) show that conformity reaches high levels at minimal peer exposure ($K{=}2$) and intensifies with greater initial diversity. Across all configurations, debate consumes 2.1-3.4$\times$ more tokens (up to 28,631 tokens per problem) than self-correction for equal or lower accuracy. Our results indicate that, within the 7-8B parameter class, homogeneous teams without structured roles do not benefit from unguided peer exchange, and that isolated self-correction consistently offers a more favorable cost-accuracy tradeoff. --- ### Leveraging Imperfect Medical Data: A Manifold-Consistent Spatio-Temporal Network for Sensor-based Human Activity Recognition Authors: Jiangtao Fan, Anish Jindal, Amir Atapour-Abarghouei Published: 2026-04-29 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00913v1 Sensor-based Human Activity Recognition (HAR) has attracted increasing attention in medical and healthcare monitoring, particularly with the growth of Internet of Medical Things (IoMT). However, in real-world wearable sensing scenarios, IoMT signals are often corrupted by missing measurements, sensor failures, and environmental noise, which significantly degrade the performance of conventional deep learning models that assume clean and complete inputs. To address this challenge, we propose a Manifold-Consistent Spatio-Temporal Network (MCSTN) for robust HAR under imperfect sensing conditions. The proposed framework introduces a dual-level corruption modeling mechanism that simulates realistic sensor imperfections through both physical-level corruption and diffusion-driven continuous corruption. By enforcing representation consistency across multiple corrupted views, the model learns stable and corruption-invariant semantic representations. Furthermore, we design a dual-stream spatio-temporal architecture that explicitly decouples temporal dynamics modeling and spatial correlation learning. The temporal stream captures long-term activity dynamics, while the spatial stream models inter-sensor relationships, enabling more effective spatio-temporal representation learning. Extensive experiments on three widely used HAR benchmark datasets, PAMAP2, Opportunity, and WISDM, demonstrate that the proposed MCSTN achieves competitive performance compared with existing state-of-the-art methods, particularly under imperfect sensing conditions. These results validate the effectiveness and robustness of the proposed framework for real-world wearable IoMT sensing applications. --- ### Object-Level Explanations for Image Geolocation Models: a GeoGuessr use-case Authors: Emilie Durrieu, Christophe Hurter, Philippe Muller, Victor Boutin Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00912v1 When humans play geolocation games such as GeoGuessr, they rely on concrete visual cues, such as road markings, vegetation, or architectural details, to infer where an image was captured. Whether image geolocation models rely on similar object-level evidence remains difficult to determine, as attribution methods like Grad-CAM typically highlight diffuse regions rather than coherent visual entities, making it difficult to link model predictions to specific objects or perceptible patterns. In this work, we propose an object-centric analysis pipeline to investigate the visual evidence used by geolocation models. Starting from attribution maps, we extract salient regions and segment them into object-like elements. We evaluate their predictive relevance through deletion and insertion tests, comparing attributionguided crops to randomly selected regions with similar coverage. Experiments on a three-country benchmark show that attribution-guided crops consistently retain more information for the model's prediction than random crops. These results suggest that attribution maps can be decomposed into interpretable, perceptible elements, providing a step toward object-level analysis of geolocation models. --- ### When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation Authors: Lin Sun, Wang Dexian, Jingang Huang, Linglin Zhang, Change Jia, Zhengwei Cheng, Xiangzheng Zhang Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00911v1 Industrial Retrieval-Augmented Generation (RAG) systems depend on optical character recognition (OCR) to transform visual documents into text. Existing OCR benchmarks rely on character-level metrics, which inadequately measure downstream RAG effectiveness under real-world conditions. We introduce an OCR benchmark for industrial RAG systems covering 11 challenging document types, including extreme layouts, high-resolution pages, complex or watermarked backgrounds, historical documents with non-standard reading orders, visually decorated text, and documents containing tables and mathematical formulas. Evaluating recent SOTA OCR models under a controlled OCR-first RAG pipeline shows clear performance degradation on realistic industrial documents despite strong conventional benchmark scores. We find that high OCR accuracy does not necessarily translate into strong downstream RAG performance: structural and semantic errors can cause substantial retrieval failures even when WER/CER remains low. Further analysis shows that this mismatch is category-dependent, arises through both retrieval-side and downstream generation-side failures, and remains stable across representative OCR-first pipeline choices. The benchmark is publicly available at https://github.com/Qihoo360/InduOCRBench. --- ### Accelerating battery research with an interoperable interface between FINALES and Kadi4Mat Authors: Giovanna Tosato, Leon Merker, Monika Vogler, Michael Selzer, Arnd Koeppe Published: 2026-04-29 Categories: cs.AI, cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2605.00909v2 This study investigates how automated and interoperable research infrastructures can accelerate experimental materials research, using battery formation as a case study. We introduce a methodological framework that integrates the FINALES and Kadi Research Data Management ecosystems, enabling coordinated experiment execution, data management, and analysis across distributed research infrastructures. The FINALES framework orchestrates experiment planning and execution on the POLiS Materials Acceleration Platform, while Kadi handles data management, visualization, experiment selection, and persistent storage. This interoperability enables reproducible, coordinated, end-to-end experimental workflows that connect automated systems and human-operated processes across multiple research sites. To showcase the framework's capabilities, we investigate the formation process of sodium-ion coin cells, a critical step that influences cell lifetime. Using a dataset generated through the proposed infrastructure, we analyze the relationships between formation parameters and electrochemical performance. A Gaussian process model is employed to identify regions associated with favorable performance within the investigated parameter domain while quantifying predictive uncertainty, highlighting where additional experimental evidence would be most informative. The presented workflow demonstrates that interoperable, automated infrastructures can support experimental campaigns, enhance data consistency and traceability, and support reproducible, data-driven analysis. While demonstrated through a battery application, the framework provides a general methodology for integrating distributed experimental platforms with RDM systems, transferable across materials science and engineering domains. --- ### Accelerating battery research with an AI interface between FINALES and Kadi4Mat Authors: Giovanna Tosato, Leon Merker, Monika Vogler, Michael Selzer, Arnd Koeppe Published: 2026-04-29 Categories: cs.AI, cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2605.00909v1 The time-consuming formation process critically impacts the longevity of sodium-ion coin cells and End Of Life (EOL) performance. This study aims to optimize formation protocols for duration efficiency, targeting high-performance outcomes while minimizing the number of experiments to reduce resource consumption and accelerate discovery. Specifically, we consider two potentially competing objectives: minimizing formation time and maximizing EOL performance. Beyond this application focus, we also present a methodological contribution: a framework designed to enable interoperability between the FINALES and Kadi RDM ecosystems, which we employ to tackle our optimization problem. In this setup, the FINALES framework orchestrates experiment planning and execution on the POLiS MAP, while an active-learning agent implemented within Kadi4Mat guides experiment selection, using multi-objective batched Bayesian optimization to efficiently explore the parameter space. This interoperability enhancement enables coordinated, distributed collaboration across automated systems and human-operated workflows, bridging multiple research centers. Using this approach, we iteratively explore the trade-off between formation time and EOL performance and identify candidate solutions approximating the Pareto front. The resulting workflow demonstrates the capability of interoperable infrastructures to facilitate data-driven optimization in battery research, and establishes a transferable framework applicable to diverse materials science and engineering optimization tasks. --- ### Evaluation of Convolutional and Transformer-Based Detectors for Weed Detection in Tomato Plantations Authors: Alcides Toledo Espinosa, Gerardo Antonio Álvarez Hernández, Ángel Eduardo Zamora-Suárez, Miguel Bolaños, Juan Irving Vásquez Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00908v2 This paper presents a comparative evaluation of convolutional and transformer-based object detection architectures for early weed detection in tomato plantations. Representative models from each paradigm are considered, including YOLOv26-nano, a recent variant of the YOLO family, and RT-DETR Large and RF-DETR Medium as transformer-based architectures. The evaluation was conducted on the GROUNDBASED_WEED dataset, considering six weed classes and an additional category corresponding to unidentified plants, which allowed for the assessment of performance in terms of detection accuracy and computational efficiency using metrics such as precision, recall, average precision, and inference speed, as well as non-parametric statistical tests. The results highlight a clear trade-off between efficiency and contextual modeling: CNN-based detectors achieve high performance at a lower computational cost, while transformer-based approaches offer better global context capture at the expense of higher resource demands. These results provide practical criteria for model selection in precision agriculture applications. --- ### TRIP-Evaluate: An Open Multimodal Benchmark for Evaluating Large Models in Transportation Authors: Han Gong, Zhen Zhou, Yunyang Shi, Yan Tan, Jinbiao Huo, Qi Hong, Zhiyuan Liu Published: 2026-04-29 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.00907v1 Large language models (LLMs) and multimodal large models (MLLMs) are increasingly used for transportation tasks such as regulation question answering, traffic management support, engineering review, and autonomous-driving scene reasoning. Yet transportation workflows are rule-intensive, computation-intensive, safety-critical, and inherently multimodal. Existing general benchmarks provide limited evidence of whether a model can apply regulations correctly, perform verifiable engineering calculations, or interpret traffic scenes reliably, while the small number of public transportation benchmarks remain narrow in scope and rarely support fine-grained diagnosis across text, images, and point-cloud data. To address this gap, we present TRIP-Evaluate, an open multimodal benchmark for large models in transportation. The benchmark organizes 837 items using a role-task-knowledge taxonomy that covers vehicle, traffic-management, traveler, and planning-and-design functions. Each item is annotated with capability, modality, and difficulty labels, enabling diagnosis from overall accuracy down to specific failure modes. The current release includes 596 text items, 198 image items, and 43 point-cloud items. TRIP-Evaluate also standardizes item construction, quality control, prompting, decoding, and scoring to improve cross-model comparability. Results on a diverse panel of models show that text-based performance is improving, but substantial weaknesses remain in multi-step engineering calculation, rule-constrained reasoning, multimodal scene understanding, and point-cloud understanding. Overall, TRIP-Evaluate provides a reproducible, diagnosable, and engineering-aligned evaluation baseline for model selection, regression testing, and safer deployment in transportation applications. --- ### Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models Authors: Hongjun Wang, Po Hu, Kai Han Published: 2026-04-29 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.00906v1 Generalized Category Discovery (GCD) aims to categorize unlabelled instances from both known and unknown classes by transferring knowledge from labelled data of known classes. Existing methods assume all data comes from a single domain, yet real-world unlabelled data often exhibits domain shifts alongside semantic shifts. We study GCD under domain shifts and propose three frameworks that adapt foundation models, ranging from self-supervised vision models to vision-language models. (i) HiLo disentangles domain and semantic features through multi-level feature extraction and mutual information minimization, combined with PatchMix augmentation and curriculum sampling. (ii) HLPrompt extends HiLo with semantic-aware spatial prompt tuning to suppress background and domain noise. (iii) VLPrompt leverages vision-language models via factorized textual prompts and cross-modal consistency regularization. The three methods share core design principles while operating on different foundation backbones, making them suitable for different deployment scenarios. Extensive experiments on synthetic corruptions and real-world multi-domain shifts demonstrate consistent improvements over strong baselines. Project page: https://visual-ai.github.io/hilo/ --- ### DIAGRAMS: A Review Framework for Reasoning-Level Attribution in Diagram QA Authors: Anirudh Iyengar Kaniyar Narayana Iyengar, Tampu Ravi Kumar, Manan Suri, Raviteja Bommireddy, Dinesh Manocha, Puneet Mathur, Vivek Gupta Published: 2026-04-29 Categories: cs.CL, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2605.00905v1 Diagram question answering (Diagram QA) requires reasoning-level attribution that links each question-answer pair to all visual regions needed to derive the answer, rather than only the region containing the final response. Creating such structured evidence across diagrams, charts, maps, circuits, and infographics is time-consuming, and existing annotation tools tightly couple their interfaces to dataset-specific formats. We present DIAGRAMS, a lightweight, schema-driven review framework that decouples interface logic from dataset-specific JSON structures through an internal meta-schema and dataset adapters. Given an image and QA pair with optional candidate regions, the system performs QA-conditioned evidence selection and proposes the regions required for reasoning. When QA pairs or candidate regions are missing, it generates them and supports human verification and refinement. Across six Diagram QA datasets, model-suggested evidence achieves 85.39% precision and 75.30% recall against reviewer-final selections (micro-averaged). These results indicate that the review-first framework reduces manual region creation while maintaining high agreement with final reasoning-level attributions. We release a public demo and installable package to support dataset auditing, grounded supervision creation, and grounded evaluation. --- ### Smart Ensemble Learning Framework for Predicting Groundwater Heavy Metal Pollution Authors: T. Ansah-Narh, G. Y. Afrifa, J. B. Tandoh, K. Asare, M. Addi, K. E. Yorke, D. M. A. Akpoley, K. Aidoo, S. K. Fosuhene Published: 2026-04-29 Categories: cs.LG, cs.AI, physics.data-an, physics.geo-ph, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2605.00056v1 Groundwater in the Densu Basin is increasingly threatened by heavy metal contamination, but conventional methods fail to capture the statistical complexity and spatial heterogeneity of pollution indicators. A key challenge is modelling the Heavy Metal Pollution Index (HPI), which is typically skewed and affected by correlated contaminants, leading to biased predictions without transformation. This study develops a predictive framework integrating response transformations with nested cross-validated ensemble machine learning. Three transformations (raw, log, and Gaussian copula) were applied to HPI and evaluated across six learners: support vector regression (SVM), $k$-nearest neighbours (k-NN), CART, Elastic Net, kernel ridge regression, and a stacked Lasso ensemble. Raw-scale models produced deceptively high fits (Elastic Net and stacked ensemble $R^2 \approx 1.0$), suggesting over-optimism. The log transformation stabilised variance (SVM: $R^2 = 0.93$, RMSE $= 0.18$; k-NN: $R^2 = 0.92$, RMSE $= 0.20$). The Gaussian copula gave the most reliable results: stacked ensemble $R^2 = 0.96$ (RMSE $= 0.19$), with other learners maintaining high accuracy. Copula-based models improved residuals and produced spatially plausible maps. DBSCAN clustering revealed Fe and Mn as primary HPI contributors, consistent with regional hydrogeochemistry. Limitations include reliance on random (not spatial) cross-validation and basin-specific scope. Future work should explore spatial validation and other geological settings. Overall, distribution-aware ensembles with clustering diagnostics offer robust, interpretable assessments of groundwater contamination. --- ### Ambient Persuasion in a Deployed AI Agent: Unauthorized Escalation Following Routine Non-Adversarial Content Exposure Authors: Diego F. Cuadros, Abdoul-Aziz Maiga Published: 2026-04-29 Categories: cs.CR, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2605.00055v1 We report a safety incident in a deployed multi-agent research system in which a primary AI agent installed 107 unauthorized software components, overwrote a system registry, overrode a prior negative decision from an oversight agent, and escalated through increasingly privileged operations up to an attempted system administrator command. The incident was preceded not by an adversarial attack but by routine content: a forwarded technology article written for human developers and shared by the principal investigator for discussion. The agent operated in a permissive environment, with unrestricted shell access, soft behavioral guidelines containing genuinely conflicting instructions, and no machine-enforced installation policy, and had recommended installing the same tool six hours earlier before being told to stand down. We analyze the behavioral cascade, the control boundaries that failed, and the limitations of multi-agent oversight in detecting and remediating the damage. We use directive weighting error as a descriptive interpretation of the observed failure and ambient persuasion as a provisional analytic label for the broader trigger configuration of non-adversarial environmental content preceding unauthorized agent action. The case highlights ethical and governance implications for deployed agent systems: ambiguous conversational cues are insufficient authorization for consequential actions, prior refusals must persist as enforceable constraints rather than message-level reminders, and oversight mechanisms require systematic post-incident auditing in addition to routine monitoring. --- ### Two-View Accumulation as the Primary Training Lever for Hybrid-Capture Gaussian Splatting: A Variance-Decomposition View of When Gradient Surgery Helps Authors: Sungjun Cho Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00052v1 Hybrid-capture novel view synthesis combines images at substantially different camera distances (e.g., aerial drone and ground-level views). Standard 3D Gaussian Splatting (3DGS), trained for 30K iterations with one rendered view per optimizer step, under-fits the minority regime by 1-3 dB on five hybrid-capture benchmarks. We isolate the lever that closes this gap. Among compute-matched alternatives -- vanilla 60K iterations, magnitude corrections (GradNorm), direction-aware near/far gradient surgery, projective preconditioning, confidence-gated sample-level surgery, and a random two-view-per-step control -- the simplest structural change wins: rendering two views per optimizer step. The pairing rule (geometry-defined near/far, random, or active loss-disparity) does not change PSNR beyond seed variance on any of the five scenes; the structural change of having two views per step does. We propose a variance-decomposition framework that predicts and explains this finding: under bimodal camera regimes, between-regime gradient variance turns out to be small relative to within-regime variance in 3DGS, so structured and random pairings are variance-equivalent in expectation, and the variance halving from two-view accumulation itself is the dominant effect. We verify the framework on five scenes whose camera-altitude bimodality coefficients span (0.55, 1.00), and we report the negative result that direction-aware projection, magnitude correction, confidence gating, and an active loss-disparity pairing all fall within seed variance of random two-view pairing. The two-view structural lever transfers cleanly to the Scaffold-GS and Pixel-GS backbones. We position this work as an honest characterization of which training-side axes do and do not move PSNR for hybrid-capture 3DGS, together with the framework that explains why. --- ### Learning from the Unseen: Generative Data Augmentation for Geometric-Semantic Accident Anticipation Authors: Yanchen Guan, Haicheng Liao, Chengyue Wang, Xingcheng Liu, Jiaxun Zhang, Keqiang Li, Zhenning Li Published: 2026-04-29 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.00051v1 Anticipating traffic accidents is a critical yet unresolved problem for autonomous driving, hindered by the inherent complexity of modeling interactions between road users and the limited availability of diverse, large-scale datasets. To address these issues, we propose a dual-path framework. On the one hand, we employ a video synthesis pipeline that, guided by structured prompts, derives feature distributions from existing corpora and produces high-fidelity synthetic driving scenes consistent with the statistical patterns of real data. On the other hand, we design a graph neural network enriched with semantic cues, enabling dynamic reasoning over both spatial and semantic relations among participants. To validate the effectiveness of our approach, we release a new benchmark dataset containing standardized, finely annotated video sequences that cover a broad spectrum of regions, weather, and traffic conditions. Evaluations across existing datasets and our new benchmark confirm notable gains in both accuracy and anticipation lead time, highlighting the capacity of the proposed framework to mitigate current data bottlenecks and enhance the reliability of autonomous driving systems. --- ### Learning physically grounded traffic accident reconstruction from public accident reports Authors: Yanchen Guan, Haicheng Liao, Chengyue Wang, Zhenning Li Published: 2026-04-29 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.00050v1 Traffic accidents are routinely documented in textual reports, yet physically grounded accident reconstruction remains difficult because detailed scene measurements and expert reconstructions are scarce, costly and hard to scale. Here we formulate accident reconstruction from publicly accessible reports and scene measurements as a parameterized multimodal learning problem. We construct CISS-REC, a dataset of 6,217 real-world accident cases curated from the NHTSA Crash Investigation Sampling System, and develop a reconstruction framework that grounds report semantics to road topology and participant attributes, reconstructs lane consistent pre-impact motion, and refines collision relevant interactions through localized geometric reasoning and temporal allocation. Our method outperforms representative baselines on CISS-REC, achieving the strongest overall reconstruction fidelity, including improved accident point accuracy and collision consistency. These results show that public accident reports can serve as scalable computational substrates for quantitatively verifiable accident reconstruction, with potential value for traffic safety analysis, simulation and autonomous driving research. --- ### SiriusHelper: An LLM Agent-Based Operations Assistant for Big Data Platforms Authors: Yu Shen, Shiyang Liu, Qihang He, Yihang Cheng, Haining Xie, Zhiming He, Huahua Fan, Xianzhi Tan, Teng Ma, Shaoquan Zhang, Danqing Huang, Fan Jiang et al. Published: 2026-04-29 Categories: cs.DB, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2605.00043v1 Big data platforms are widely used in modern enterprises, and an in-production intelligent assistant is increasingly important to help users quickly find actionable guidance and reduce operational burden. While recent LLM+RAG assistants provide a natural interface, they face practical challenges in real deployments: limited scenario coverage across both general consultation and domain-specific troubleshooting workflows, inefficient knowledge access due to inadequate multi-hop retrieval and flat knowledge organization, and high maintenance cost because escalated tickets are unstructured and hard to convert into assistant improvements and reusable SOPs. In this paper, we present SiriusHelper, a deployed intelligent assistant for big data platforms. SiriusHelper serves as a unified online assistant that automatically identifies user intent and routes queries to the right handling path, including dedicated expert workflows for specialized scenarios (e.g., SQL execution diagnosis). To support complex troubleshooting, SiriusHelper combines a DeepSearch-driven mechanism with a priority-based hierarchical knowledge base to enable multi-hop retrieval without context overload, thus improving answer reliability and latency. To reduce expert overhead, SiriusHelper further introduces automated ticket understanding and SOP distillation: it diagnoses the assistant failure reason (e.g., missing knowledge or wrong routing) and extracts domain-specific SOPs to continuously enrich the knowledge base. Experiments and online deployment on Tencent Big Data platform show that SiriusHelper outperforms representative alternatives and reduces online ticket volume by 20.8\%. --- ### AdvDMD: Adversarial Reward Meets DMD For High-Quality Few-Step Generation Authors: Xu Wang, Zexian Li, Litong Gong, Tiezheng Ge, Zhijie Deng Published: 2026-04-29 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.28126v1 Diffusion models offer superior generation quality at the expense of extensive sampling steps. Distillation methods, with Distribution Matching Distillation (DMD) as a popular example, can mitigate this issue, but performance degradation remains pronounced when sampling steps are limited. Reinforcement learning (RL) has been leveraged to improve the few-step generation quality during distillation, with the potential to even surpass the performance of the teacher model. However, existing approaches are combinatorial in nature, merely integrating an RL process with the distillation process, which introduces unnecessary complexities. To address this gap, we propose AdvDMD, a method that seamlessly unifies DMD distillation and RL. Specifically, AdvDMD employs the adversarially trained discriminator from DMD2 as the reward model, which assigns low scores to generated images and high scores to real ones. It is trained on both intermediate and final states of the denoising process and updated online with the distilled model, enabling a holistic supervision of the sampling trajectories and mitigating reward hacking. We adopt a unified SDE backward simulation and a different training schedule for DMD and RL to enable a more stable and efficient training. Experimental results demonstrate that the 4-step AdvDMD outperforms the original 40-step model for SD3.5 on DPG-Bench, while achieving significant performance gains for SD3 on the GenEval. On Qwen-Image, our 2-step AdvDMD achieves superior performance over TwinFlow. --- ### When 2D Tasks Meet 1D Serialization: On Serialization Friction in Structured Tasks Authors: Chung-Hsiang Lo, Lu Li, Diji Yang, Tianyu Zhang, Yunkai Zhang, Yoshua Bengio, Yi Zhang Published: 2026-04-29 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.27272v2 In the LLM era, many symbolic and structured problems are presented to models through 1D text serialization. Yet some such problems are natively two-dimensional: their relevant relations, such as row--column correspondence or spatial adjacency, are defined by position in a 2D layout rather than by sequential order. This raises a representational question: does preserving the same symbolic entries in a 1D sequence also preserve the relational structure needed for computation? We study this issue through the lens of serialization friction: the representational mismatch in which the same underlying task instances and entries are still present, but relations that depend on layout become implicit under 1D serialization. The study uses a controlled synthetic testbed of three tasks: matrix transpose, Conway's Game of Life, and LU decomposition. In each task, the same instances are presented either as 1D text serialization or as their native 2D layout rendered as an image. Across this testbed, 1D serialization degrades more sharply as task size grows, and errors under serialization exhibit spatially structured patterns, suggesting that this presentation choice is consequential within our testbed. To further interpret these results, we add supplementary analyses that include a within-visual probe and an additional comparison of the two input presentations under the mixed-training transpose setting. These findings suggest that, for layout-defined tasks, reducing inputs to 1D serialization is not a neutral choice of representation. --- ### Unifying biomedical knowledge in a modern multimodal graph Authors: Lucas Vittor, Ayush Noori, Iñaki Arango, Joaquín Polonuer, Sam Rodriques, Andrew White, David A. Clifton, Marinka Zitnik Published: 2026-04-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.27269v3 Biomedical knowledge graphs (KGs) are widely used in the life sciences, yet many are derived from unstructured documents and therefore lack schema-level constraints, whereas graphs assembled from structured resources are difficult to harmonize into a unified representation. We present OptimusKG, a multimodal biomedical labeled property graph (LPG) built from structured and semi-structured resources to preserve factual, type-specific metadata across molecular, anatomical, clinical, and environmental domains. OptimusKG contains 190,939 nodes across 10 entity types, 21,818,752 edges across 27 edge types, and 67,070,490 property instances encoding 109,665,797 values across 145 distinct property keys, derived from 18 ontologies and controlled vocabularies. The graph enforces a top-level schema for nodes and edges and retains granular, type-specific properties, cross-references, and provenance. We assessed the validity of OptimusKG by evaluating whether graph relationships are supported by evidence from the scientific literature using a multimodal agent, PaperQA3. PaperQA3 identified supporting evidence for 70.0% of sampled edges, whereas 83.4% of sampled false edges received no supporting evidence. Edges without literature support were concentrated in associations derived from experimental and functional genomics resources, suggesting that OptimusKG captures biomedical knowledge that may precede synthesis in the scientific literature. OptimusKG is distributed as Apache Parquet files, providing a standardized resource for graph-based machine learning, knowledge-grounded retrieval with large language models, and biomedical discovery use cases such as hypothesis generation. --- ### Unifying biomedical knowledge in a modern multimodal graph Authors: Lucas Vittor, Ayush Noori, Iñaki Arango, Joaquín Polonuer, Sam Rodriques, Andrew White, David A. Clifton, Marinka Zitnik Published: 2026-04-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.27269v2 Biomedical knowledge graphs (KGs) are widely used in the life sciences, yet many are derived from unstructured documents and therefore lack schema-level constraints, whereas graphs assembled from structured resources are difficult to harmonize into a unified representation. We present OptimusKG, a multimodal biomedical labeled property graph (LPG) built from structured and semi-structured resources to preserve factual, type-specific metadata across molecular, anatomical, clinical, and environmental domains. OptimusKG contains 190,939 nodes across 10 entity types, 21,818,752 edges across 27 edge types, and 67,070,490 property instances encoding 109,665,797 values across 145 distinct property keys, derived from 18 ontologies and controlled vocabularies. The graph enforces a top-level schema for nodes and edges and retains granular, type-specific properties, cross-references, and provenance. We assessed the validity of OptimusKG by evaluating whether graph relationships are supported by evidence from the scientific literature using a multimodal agent, PaperQA3. PaperQA3 identified supporting evidence for 70.0% of sampled edges, whereas 83.4% of sampled false edges received no supporting evidence. Edges without literature support were concentrated in associations derived from experimental and functional genomics resources, suggesting that OptimusKG captures biomedical knowledge that may precede synthesis in the scientific literature. OptimusKG is distributed as Apache Parquet files, providing a standardized resource for graph-based machine learning, knowledge-grounded retrieval with large language models, and biomedical discovery use cases such as hypothesis generation. --- ### From Prompt to Physical Actuation: Holistic Threat Modeling of LLM-Enabled Robotic Systems Authors: Neha Nagaraja, Hayretdin Bahsi, Carlo R. da Cunha Published: 2026-04-29 Categories: cs.CR, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2604.27267v2 As large language models are integrated into autonomous robotic systems for task planning and control, compromised inputs or unsafe model outputs can propagate through the planning pipeline to physical-world consequences. Although prior work has studied robotic cybersecurity, adversarial perception attacks, and LLM safety independently, no existing study traces how these threat categories interact and propagate across trust boundaries in a unified architectural model. We address this gap by modeling an LLM-enabled autonomous robot in an edge-cloud architecture as a hierarchical Data Flow Diagram and applying STRIDE-per-interaction analysis across six boundary-crossing interaction points using a three-category taxonomy of Conventional Cyber Threats, Adversarial Threats, and Conversational Threats. The analysis reveals that these categories converge at the same boundary crossings, and we trace three cross-boundary attack chains from external entry points to unsafe physical actuation, each exposing a distinct architectural property: the absence of independent semantic validation between user input and actuator dispatch, cross-modal translation from visual perception to language-model instruction, and unmediated boundary crossing through provider-side tool use. To our knowledge, this is the first DFD-based threat analysis integrating all three threat categories across the full perception-planning-actuation pipeline of an LLM-enabled robotic system. --- ### AutoREC: A reinforcement learning platform for equivalent circuit model generation Authors: Ali Jaberi, Yonatan Kurniawan, Robert Black, Shayan Mousavi M., Kabir Verma, Zoya Sadighi, Santiago Miret, Jason Hattrick-Simpers Published: 2026-04-29 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2604.27266v2 This paper introduces AutoREC, an open-source Python platform for developing, training, and evaluating reinforcement learning (RL) agents that automatically generate equivalent circuit models (ECMs) from electrochemical impedance spectroscopy (EIS) data. Although ECMs are widely used to interpret EIS measurements, their identification typically relies on manual trial-and-error, requiring domain expertise and limiting scalability, particularly in autonomous experimental pipelines such as self-driving laboratories. In AutoREC, ECM generation is formulated as a Markov decision process in which an RL agent sequentially modifies a circuit topology based on the current state, available actions, and feedback from the resulting model. The platform supports an end-to-end workflow encompassing EIS preprocessing with selectable impedance representations, agent setup and training, ECM generation for new measurements, and visualization-based evaluation and analysis of agent decision-making. AutoREC implements a configurable Double Deep Q-Network (DDQN) agent with prioritized experience replay and a dedicated dead-loop mitigation strategy for navigating the complex circuit-generation action space efficiently. To demonstrate the platform, we trained and evaluated a representative agent on synthetic EIS datasets and applied it to previously unseen experimental spectra from battery, corrosion, oxygen evolution reaction, and CO$_2$ reduction systems. These case studies illustrate the end-to-end capabilities of AutoREC while revealing challenges associated with experimental complexity and limited training-data coverage. The demonstrated agent serves as a reference implementation; AutoREC provides an extensible foundation through which users can develop and evaluate agents tailored to their specific electrochemical systems and research objectives. --- ### AutoREC: A software platform for developing reinforcement learning agents for equivalent circuit model generation from electrochemical impedance spectroscopy data Authors: Ali Jaberi, Yonatan Kurniawan, Robert Black, Shayan Mousavi M., Kabir Verma, Zoya Sadighi, Santiago Miret, Jason Hattrick-Simpers Published: 2026-04-29 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2604.27266v1 This paper introduces AutoREC, an open-source Python package for developing reinforcement learning (RL) agents to automatically generate equivalent circuit models (ECMs) from electrochemical impedance spectroscopy (EIS) data. While ECMs are a standard framework for interpreting EIS data, traditional identification is typically based on manual trial-and-error, which requires domain experts and limits scalability, particularly in autonomous experimental pipelines such as self-driving laboratories. AutoREC addresses this challenge by formulating ECM construction as a sequential decision-making problem within a Markov Decision Process framework. It implements a Double Deep Q-Network with prioritized experience replay, along with a dedicated dead-loop mitigation strategy, to efficiently explore a complex action space for circuit generation. To demonstrate the capabilities of the platform, we trained an RL agent using AutoREC and evaluated its strengths and limitations across diverse datasets, while also discussing possible strategies to mitigate these limitations in future agent designs. The trained agent achieved a success rate exceeding $99.6\%$ on synthetic datasets and demonstrated strong generalization to unseen experimental EIS data from batteries, corrosion, oxygen evolution reaction, and CO$_2$ reduction systems. These results position AutoREC as a promising platform for adaptive and data-driven ECM generation, with potential for integration into automated electrochemical workflows. --- ### Self-Evolving Software Agents Authors: Marco Robol, Paolo Giorgini Published: 2026-04-29 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.27264v1 Autonomous agents can adapt their behaviour to changing environments, but remain bound to requirements, goals, and capabilities fixed at design time, preventing genuine software evolution. This paper introduces self-evolving software agents, combining BDI reasoning with LLMs to enable autonomous evolution of goals, reasoning, and executable code. We propose a BDI-LLM architecture in which an automated evolution module operates alongside the agent's reasoning loop, eliciting new requirements from experience and synthesizing corresponding design and code updates. A prototype evaluated in a dynamic multi-agent environment shows that agents can autonomously discover new goals and generate executable behaviours from minimal prior knowledge. The results indicate both the feasibility and current limits of LLM-driven evolution, particularly in terms of behavioural inheritance and stability. --- ### Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation Authors: Théo Gigant, Bowen Peng, Jeffrey Quesnelle Published: 2026-04-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.27263v2 Subword tokenization is an essential part of modern large language models (LLMs), yet its specific contributions to training efficiency and model performance remain poorly understood. In this work, we decouple the effects of subword tokenization by isolating them within a controlled byte-level pretraining pipeline. We formulate and test hypotheses across various dimensions, including sample throughput, vocabulary scaling, and the linguistic prior of subword boundaries. By simulating these effects in a byte-level setting, we refine our understanding of why subword models outperform raw byte models and offer insights to improve the pretraining of future byte-level and subword models. Specifically, our experiments highlight the critical role of increased training throughput and the integration of subword boundaries as either explicit priors or inductive biases. --- ### VTBench: A Multimodal Framework for Time-Series Classification with Chart-Based Representations Authors: Madhumitha Venkatesan, Xuyang Chen, Dongyu Liu Published: 2026-04-29 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.27259v1 Time-series classification (TSC) has advanced significantly with deep learning, yet most models rely solely on raw numerical inputs, overlooking alternative representations. While texture-based encodings such as Gramian Angular Fields (GAF) and Recurrence Plots (RP) convert time series into 2D images, they often require heavy preprocessing and yield less intuitive representations. In contrast, chart-based visualizations offer more interpretable alternatives and show promise in specific domains; however, their effectiveness remains underexplored, with limited systematic evaluation across chart types, visual encoding choices, and datasets. In this work, we introduce VTBench, a systematic and extensible framework that re-examines TSC through multimodal fusion of raw sequences and chart-based visualizations. VTBench generates lightweight, human-interpretable plots -- line, area, bar, and scatter, providing complementary views of the same signal. We develop a modular architecture supporting multiple fusion strategies, including single-chart visual-numerical fusion, multi-chart visual fusion, and full multimodal fusion with raw inputs. Through experiments across 31 UCR datasets, we show that: (1) chart-only models are competitive in selected settings, particularly on smaller datasets; (2) combining multiple chart types can improve accuracy by capturing complementary visual cues; and (3) multimodal models improve or maintain performance when visual features provide non-redundant information, but may degrade accuracy when they introduce redundancy. We further distill practical guidelines for selecting chart types, fusion strategies, and configurations. VTBench establishes a unified foundation for interpretable and effective multimodal time-series classification. --- ### Towards Accelerated SCF Workflows with Equivariant Density-Matrix Learning and Analytic Refinement Authors: Zuriel Y. Yescas-Ramos, Andrés Álvarez-García, Huziel E. Sauceda Published: 2026-04-29 Categories: physics.chem-ph, cs.AI, cs.LG, physics.comp-ph, quant-ph arXiv: https://arxiv.org/abs/2604.27256v1 We present \textsc{dm-PhiSNet}, a physically constrained \textsc{PhiSNet}-based equivariant model that predicts one-electron reduced density matrices (1-RDMs) directly from molecular geometries in an atomic-orbital (AO) basis for accelerated self-consistent field (SCF) workflows. Training follows a two-stage schedule with progressively introduced physically motivated objectives, and the resulting predictions are refined by a lightweight analytic block. This block enforces electron-number conservation, drives the 1-RDM toward generalized idempotency in the AO metric, and regularizes the occupation spectrum of the Löwdin-orthogonalized density. Across six closed-shell systems -- H$_2$O, CH$_4$, NH$_3$, HF, ethanol, and NO$_3^-$ -- the refined 1-RDMs provide SCF initial guesses that substantially reduce iteration steps by 49--81\% relative to standard initializations. Beyond SCF acceleration, the learned 1-RDMs yield accurate one-shot total energies and Hellmann--Feynman atomic forces without force supervision, indicating that the model captures chemically meaningful electronic structure. These results demonstrate that combining equivariant learning with analytic constraint enforcement provides a simple, general route to solver-ready density-matrix initializations and accelerated SCF workflows. --- ### AutoSurfer -- Teaching Web Agents through Comprehensive Surfing, Learning, and Modeling Authors: Fazle Elahi Faisal, Qianhui Wu, Baolin Peng, Jianfeng Gao Published: 2026-04-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.27253v1 Recent advances in multimodal large language models (LLMs) have revolutionized web agents that can automate complex tasks on websites. However, their accuracy remains limited by the scarcity of high-quality web trajectory training data. Existing automatic trajectory generation methods suffer from incomplete website coverage due to homepage-based task proposals or random-walk exploration. Such methods often result in hallucinated or ambiguous task synthesis that lead to incomplete and unreliable trajectory generation. Here, we present AutoSurfer, a comprehensive web trajectory generator that addresses these limitations through three key innovations. First, AutoSurfer employs a systematic breadth-first exploration strategy that maintains a queue of discovered pages and action traces, propagates knowledge across pages to avoid redundant exploration, and recursively expands multi-level graphical user interface elements - closely resembling how a human would learn a new website. Second, AutoSurfer leverages the exploration trajectory to guide task synthesis, reducing hallucinations by grounding complex tasks in actual navigation paths rather than isolated actions or page content alone. Third, AutoSurfer uses the same exploration trajectory as hints to steer a web agent toward more accurate and reliable trajectory refinement. Together, these innovations enable AutoSurfer to comprehensively cover a website's action space and generate data suitable for training website-specific LLMs. We evaluate AutoSurfer on the WebArena benchmark by fine-tuning Qwen2.5-VL-7B-Instruct and demonstrate that it outperforms state-of-the-art methods - Explorer, OS-Genesis, and SynthAgent - achieving up to 24.23% overall task completion accuracy compared to 19.59% for the best prior method. Further, task diversity analysis demonstrates that AutoSurfer yields a more diverse distribution of synthesized tasks. --- ### Compliance vs. Sensibility: On the Reasoning Controllability in Large Language Models Authors: Xingwei Tan, Marco Valentino, Mahmud Elahi Akhter, Yuxiang Zhou, Maria Liakata, Nikolaos Aletras Published: 2026-04-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.27251v3 Large Language Models (LLMs) acquire reasoning capabilities through shared inference patterns in pre-training data, which are further elicited via Chain-of-Thought (CoT). However, whether fundamental reasoning patterns, such as induction, deduction, and abduction, can be decoupled from specific problem instances remains a critical challenge for model controllability. In this paper, we present the first systematic investigation of this problem through the lens of reasoning conflicts, an explicit tension between parametric and contextual information induced by mandating logical schemata that deviate from those expected for a target task. Our evaluation reveals that LLMs consistently prioritize sensibility over compliance, favoring task-appropriate reasoning patterns despite conflicting instructions. We further demonstrate that reasoning conflicts are internally detectable, as confidence scores drop during conflicting episodes. Probing reveals instruction decodability even without compliance, while CKA identifies geometric differences across models and spans. Leveraging these insights, we steer models towards compliance, increasing instruction following by up to 29%. Overall, our findings establish that while LLM reasoning is anchored to concrete instances, active mechanistic interventions can effectively decouple logical schemata from data, offering a path toward improved controllability, faithfulness, and generalizability. --- ### Compliance versus Sensibility: On the Reasoning Controllability in Large Language Models Authors: Xingwei Tan, Marco Valentino, Mahmud Elahi Akhter, Yuxiang Zhou, Maria Liakata, Nikolaos Aletras Published: 2026-04-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.27251v2 Large Language Models (LLMs) are known to acquire reasoning capabilities through shared inference patterns in pre-training data, which are further elicited via Chain-of-Thought (CoT) practices. However, whether fundamental reasoning patterns, such as induction, deduction, and abduction, can be decoupled from specific problem instances remains a critical challenge for model controllability, and for shedding light on reasoning controllability. In this paper, we present the first systematic investigation of this problem through the lens of reasoning conflicts: an explicit tension between parametric and contextual information induced by mandating logical schemata that deviate from those expected for a target task. Our evaluation reveals that LLMs consistently prioritize sensibility over compliance, favoring task-appropriate reasoning patterns despite conflicting instructions. We further demonstrate that reasoning conflicts are internally detectable, as confidence scores significantly drop during conflicting episodes. Probing experiments confirm that reasoning types are linearly encoded from middle-to-late layers, indicating the potential for activation-level controllability. Leveraging these insights, we steer models towards compliance, increasing instruction following by up to 29%. Overall, our findings establish that while LLM reasoning is anchored to concrete instances, active mechanistic interventions can effectively decouple logical schemata from data, offering a path toward improved controllability, faithfulness, and generalizability. --- ### Instruction Complexity Induces Positional Collapse in Adversarial LLM Evaluation Authors: Jon-Paul Cacioli Published: 2026-04-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.27249v1 When instructed to underperform on multiple-choice evaluations, do language models engage with question content or fall back on positional shortcuts? We map the boundary between these regimes using a six-condition adversarial instruction-specificity gradient administered to two instruction-tuned LLMs (Llama-3-8B and Llama-3.1-8B) on 2,000 MMLU-Pro items. Distributional screening (response-position entropy) and an independent content-engagement criterion (difficulty-accuracy correlation) jointly characterise each condition. The gradient reveals three regimes rather than a monotonic transition. Vague adversarial instructions produce moderate accuracy reduction with preserved content engagement. Standard sandbagging and capability-imitation instructions produce positional entropy collapse with partial content engagement. A two-step answer-aware avoidance instruction produces extreme positional collapse, with near-total concentration on a single response position (99.9% and 87.4%) and no measurable content sensitivity. This was the only multi-step instruction tested, and it produced the most extreme shortcut. The attractor position matches each model's content-absent null-prompt default. The effect replicates across both models and four academic domains. Distributional collapse and content engagement can co-occur (50% concordance between screening criteria), indicating that entropy-based screening and difficulty-based content assessment capture partially independent dimensions of response validity. Results suggest that instruction complexity can determine whether adversarial compliance uses content-aware or content-blind mechanisms in small instruction-tuned LLMs under greedy decoding. --- ### Towards Generalizable Mapping of Hedges and Linear Woody Features from Earth Observation Data: a national Product for Germany Authors: Thorsten Hoeser, Verena Huber-Garcia, Sarah Asam, Ursula Gessner, Claudia Kuenzer Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.27247v2 Hedges and other linear woody features provide valuable ecosystem services, particularly within intensively managed agricultural landscapes. They are key elements for climate adaptation and biodiversity amongst others not only due to a largely varying flora, but also as a feeding-, resting-, and nesting place for many animals and insects including valuable pollinators. Therefore, they require dedicated management, preservation, and attention. Thus, systematic and large-scale mapping of these features from Earth observation data is of high importance. However, transferable and reusable workflows for linear woody feature mapping remain a key methodological challenge, given the diversity of sensor types, spatial resolutions, data acquisition conditions, and complex landscape variability encountered across study areas. We introduce a modular workflow built around two independently optimizable components. Firstly, a flexible input data interface that consolidates heterogeneous Earth observation data into a binary woody vegetation mask, and secondly, a deep neural network trained to separate linear from non-linear shapes within these masks. We demonstrate the workflow by deriving three national-scale linear woody feature maps for all of Germany from three input sources with 0.73 m, 1 m and 3 m spatial resolution, respectively, by using a single trained model without retraining. Evaluation against refined reference data from four federal state biotope mapping campaigns and comparison with two existing linear woody feature maps demonstrate that the workflow produces competitive results across all evaluation sites on a national level. The modular design and its demonstrated applicability at national scale provide a foundation for scalable and generalizable linear woody feature mapping beyond Germany. --- ### Addressing the Reality Gap: A Three-Tension Framework for Agentic AI Adoption Authors: Jason Fournier, Kacper Łodzikowski Published: 2026-04-29 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2604.27245v2 Generative AI has rapidly entered education through free consumer tools, outpacing the ability of schools and universities to respond. Now a new wave of more autonomous agentic AI systems--with the capacity to plan and act towards goals--promises both greater educational personalization and greater disruption. This chapter argues that successfully navigating these innovations requires balancing three core tensions: (1) Implementation Feasibility, or the practical capacity to integrate AI sustainably into real classrooms; (2) Adaptation Speed, or the mismatch between fast-evolving AI capabilities and the slower pace of educational change; and (3) Mission Alignment, or the need to ensure AI applications uphold educational values such as equity, privacy, and pedagogical integrity. First, we review early evidence of generative and agentic AI in various sectors and in frontline education to illustrate these tensions in context. Then, we present a three-tension framework to guide decision-makers in evaluating and designing AI initiatives across K-12 and higher education. We provide examples of how the framework can be applied to plan responsible AI deployments, and we identify emerging trends--such as curriculum-linked AI agents and educator-informed AI design--along with open research directions. We conclude the chapter with recommendations for educational leaders to proactively engage with the opportunities and challenges of AI, so that this technology can be harnessed to enhance teaching and learning in the decade ahead. --- ### Analytical Correction for Subsampling Bias in Drifting Models Authors: Jiaru Zhang, Zeyun Deng, Juanwu Lu, Ziran Wang, Ruqi Zhang Published: 2026-04-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.27239v1 Drifting models are capable one-step generative models trained to follow a drifting field. The field combines attractive and repulsive softmax-weighted centroids over the data and current-generator distributions. In practice, only a minibatch of $n$ samples from each distribution is available, and each centroid is approximated by an empirical estimate. In this paper, we begin by showing that the minibatch centroid is in general a biased estimator of the target centroid, with a pointwise $O(1/n)$ bias arising from softmax self-normalization. Correcting this bias requires the expectation over the full distribution, which is intractable. We instead approximate the leading bias term from in-batch statistics and propose Analytical Bias Correction (ABC), a closed-form plug-in adjustment. We prove that ABC reduces the bias from $O(1/n)$ to $O(1/n^2)$, introduces no first-order increase in total variance, and preserves convex-hull containment of the corrected centroid. In practice, ABC requires only two additional lines of code and has negligible wall-time overhead under compiled execution. Toy experiments confirm the theoretical $O(1/n)$ and $O(1/n^2)$ scaling. On CIFAR-10, ABC reduces FID and trains faster, with the largest gains at small $n$, where the bias is most significant. --- ### Remaining Useful Life Estimation for Turbofan Engines: A Comparative Study of Classical, CNN, and LSTM Approaches Authors: Astitva Goel, Samarth Galchar, Sumit Kanu Published: 2026-04-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.27234v1 Remaining Useful Life (RUL) estimation is a critical component of Prognostics and Health Management (PHM), enabling proactive maintenance scheduling and reducing unplanned failures in industrial equipment. This paper presents a comparative study of machine learning approaches for RUL estimation on the NASA C-MAPSS turbofan engine dataset: classical baselines (Ridge Regression, Polynomial Ridge, and XGBoost), a 1D Convolutional Neural Network (CNN), and a Long Short-Term Memory (LSTM) network. All models are evaluated on the FD001 and FD003 subsets under an identical preprocessing pipeline to ensure a fair comparison. Among raw-sequence models, the LSTM achieves RMSE of 14.93 and 14.20 on FD001 and FD003 respectively, outperforming the deep LSTM reported by Zheng et al.~\cite{paper} (RMSE 16.14 and 16.18) despite using a simpler single-layer architecture. The 1D CNN achieves RMSE of 16.97 on FD001 and 15.68 on FD003, demonstrating competitive performance on FD003 while producing more conservative RUL predictions on FD001. Ridge Regression is evaluated on raw and engineered features, while other classical models use only engineered inputs. XGBoost achieves an RMSE of 13.36 on FD003, highlighting the competitiveness of nonlinear modeling. --- ### Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents Authors: Anh Ta, Junjie Zhu, Shahin Shayandeh Published: 2026-04-29 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2604.27233v1 Tool-calling agents are evaluated on tool selection, parameter accuracy, and scope recognition, yet LLM trajectory assessments remain inherently post-hoc. Disconnected from the active execution loop, such assessments identify errors that are usually addressed through prompt-tuning or retraining, and fundamentally cannot course-correct the agent in real time. To close this gap, we move evaluation into the execution loop at inference time: a specialized reviewer agent evaluates provisional tool calls prior to execution, shifting the paradigm from post-hoc recovery to proactive evaluation and error mitigation. In practice, this architecture establishes a clear separation of concerns between the primary execution agent and a secondary review agent. As with any multi-agent system, the reviewer can introduce new errors while correcting others, yet no prior work to our knowledge has systematically measured this tradeoff. To quantify this tradeoff, we introduce Helpfulness-Harmfulness metrics: helpfulness measures the percentage of base agent errors that feedback corrects; harmfulness measures the percentage of correct responses that feedback degrades. These metrics directly inform reviewer design by revealing whether a given model or prompt provides net positive value. We evaluate our approach on BFCL (single-turn) and Tau2-Bench (multi-turn stateful scenarios), achieving +5.5% on irrelevance detection and +7.1% on multi-turn tasks. Our metrics reveal that reviewer model choice is critical: the reasoning model o3-mini achieves a 3:1 benefit-to-risk ratio versus 2.1:1 for GPT-4o. Automated prompt optimization via GEPA provides an additional +1.5-2.8%. Together, these results demonstrate a core advantage of separating execution and review: the reviewer can be systematically improved through model selection and prompt optimization, without retraining the base agent. --- ### Targeted Linguistic Analysis of Sign Language Models with Minimal Translation Pairs Authors: Serpil Karabüklü, Kanishka Misra, Shester Gueuwou, Diane Brentari, Greg Shakhnarovich, Karen Livescu Published: 2026-04-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.27232v3 Models of sign language have historically lagged behind those for spoken language (text and speech). Recent work has greatly improved their performance on tasks like sign language translation and isolated sign recognition. However, it remains unclear to what extent existing models capture various linguistic phenomena of sign language, and how well they use cues from the multiple articulators used in sign language (hands, upper body, face). We introduce a new benchmark dataset for American Sign Language, ASL Minimal Translation Pairs (ASL-MTP), divided into multiple types of sign language phenomena and corresponding minimal pairs of translations, for performing such linguistic analyses. As a case study, we use ASL-MTP to analyze a state-of-the-art ASL-to-English translation model. We conduct a targeted analysis of the model by ablating various input cues during training and inference and evaluating on the phenomena in ASL-MTP. Our results show that, while the model performs above chance level on most of the phenomena, it relies strongly on manual cues while often missing crucial non-manual cues. --- ### Upskilling with Generative AI: Practices and Challenges for Freelance Knowledge Workers Authors: Kashif Imteyaz, Isabel Lopez, Nakul Rajpal, Hunjun Shin, Saiph Savage Published: 2026-04-29 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2604.27231v1 Freelance workers must continually acquire new skills to remain competitive in online labor markets, yet they lack the organizational training, mentorship, and infrastructure available to traditional employees. Generative AI-powered tools like ChatGPT are reshaping market skill demands while also offering new forms of on-demand learning support to meet those demands. Despite growing interest in AI-powered learning tools, little is known about how freelancers actually use these tools to learn, the challenges they encounter, and how generative AI for learning interacts with precarity and competition in platform-based work. We present a mixed-methods study combining a survey and semi-structured interviews with freelance knowledge workers. Grounded in self-directed learning theory, we examine how freelancers integrate generative AI tools into their learning practices. Our findings show that freelancers increasingly rely on generative AI to structure learning and support exploratory skill acquisition, but do not treat it as their primary learning resource due to inconsistency, lack of contextual relevance, and verification overhead. We identify a shift from learning as growth to learning as survival, where upskilling is oriented toward immediate market viability rather than long-term development. We also surface a structural challenge we term invisible competencies, in which workers acquire skills through generative AI tools but lack credible ways to signal or validate these skills in competitive freelance markets. Based on these insights, we offer design recommendations for generative AI-powered learning tools for freelancers. --- ### When Roles Fail: Epistemic Constraints on Advocate Role Fidelity in LLM-Based Political Statement Analysis Authors: Juergen Dietrich Published: 2026-04-29 Categories: cs.AI, cs.CL, cs.CY, cs.MA arXiv: https://arxiv.org/abs/2604.27228v1 Democratic discourse analysis systems increasingly rely on multi-agent LLM pipelines in which distinct evaluator models are assigned adversarial roles to generate structured, multi-perspective assessments of political statements. A core assumption is that models will reliably maintain their assigned roles. This paper provides the first systematic empirical test of that assumption using the TRUST pipeline. We develop an epistemic stance classifier that identifies advocate roles from reasoning text without relying on surface vocabulary, and measure role fidelity across 60 political statements (30 English, 30 German) using four metrics: Role Drift Index (RDI), Expected Drift Distance (EDD), Directional Drift Index (DDI), and Entropy-based Role Stability (ERS). We identify two failure modes - the Epistemic Floor Effect (fact-check results create an absolute lower bound below which the legitimizing role cannot be maintained) and Role-Prior Conflict (training-time knowledge overrides role instructions for factually unambiguous statements) - as manifestations of a single mechanism: Epistemic Role Override (ERO). Model choice significantly affects role fidelity: Mistral Large outperforms Claude Sonnet by 28pp (67% vs. 39%) and exhibits a qualitatively different failure mode - role abandonment without polarity reversal - compared to Claude's active switch to the opposing stance. Role fidelity is language-robust. Fact-check provider choice is not universally neutral: Perplexity significantly reduces Claude's role fidelity on German statements (Delta = -15pp, p = 0.007) while leaving Mistral unaffected. These findings have direct implications for multi-agent LLM validation: a system validated without role fidelity measurement may systematically misrepresent the epistemic diversity it was designed to provide. --- ### Web2BigTable: A Bi-Level Multi-Agent LLM System for Internet-Scale Information Search and Extraction Authors: Yuxuan Huang, Yihang Chen, Zhiyuan He, Yuxiang Chen, Ka Yiu Lee, Huichi Zhou, Weilin Luo, Meng Fang, Jun Wang Published: 2026-04-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.27221v1 Agentic web search increasingly faces two distinct demands: deep reasoning over a single target, and structured aggregation across many entities and heterogeneous sources. Current systems struggle on both fronts. Breadth-oriented tasks demand schema-aligned outputs with wide coverage and cross-entity consistency, while depth-oriented tasks require coherent reasoning over long, branching search trajectories. We introduce \textbf{Web2BigTable}, a multi-agent framework for web-to-table search that supports both regimes. Web2BigTable adopts a bi-level architecture in which an upper-level orchestrator decomposes the task into sub-problems and lower-level worker agents solve them in parallel. Through a closed-loop run--verify--reflect process, the framework jointly improves decomposition and execution over time via persistent, human-readable external memory, with self-evolving updates to each single-agent. During execution, workers coordinate through a shared workspace that makes partial findings visible, allowing them to reduce redundant exploration, reconcile conflicting evidence, and adapt to emerging coverage gaps. Web2BigTable sets a new state of the art on WideSearch, reaching an Avg@4 Success Rate of \textbf{38.50} ($7.5\times$ the second best at 5.10), Row F1 of \textbf{63.53} (+25.03 over the second best), and Item F1 of \textbf{80.12} (+14.42 over the second best). It also generalises to depth-oriented search on XBench-DeepSearch, achieving 73.0 accuracy. Code is available at https://github.com/web2bigtable/web2bigtable. --- ### AttriBE: Quantifying Attribute Expressivity in Body Embeddings for Recognition and Identification Authors: Basudha Pal, Siyuan Huang, Anirudh Nanduri, Zhaoyang Wang, Rama Chellappa Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.27218v1 Person re-identification (ReID) systems that match individuals across images or video frames are essential in many real-world applications. However, existing methods are often influenced by attributes such as gender, pose, and body mass index (BMI), which vary in unconstrained settings and raise concerns related to fairness and generalization. To address this, we extend the notion of expressivity, defined as the mutual information between learned features and specific attributes, using a secondary neural network to quantify how strongly attributes are encoded. Applying this framework to three transformer-based ReID models on a large-scale visible-spectrum dataset, we find that BMI consistently shows the highest expressivity in deeper layers. Attributes in the final representation are ranked as BMI > Pitch > Gender > Yaw, and expressivity evolves across layers and training epochs, with pose peaking in intermediate layers and BMI strengthening with depth. We further extend the analysis to cross-spectral person identification across infrared modalities including short-wave, medium-wave, and long-wave infrared. In this setting, pitch becomes comparable to BMI and attribute trends increase monotonically across depth, suggesting increased reliance on structural cues when bridging modality gaps. Overall, the results show that transformer-based ReID embeddings encode a hierarchy of implicit attributes, with morphometric information persistently embedded and pose contributing more strongly under cross-spectral conditions. --- ### Toward Personalized Digital Twins for Cognitive Decline Assessment: A Multimodal, Uncertainty-Aware Framework Authors: Bulent Soykan, Gulsah Hancerliogullari Koksalmis, Hsin-Hsiung Huang, Laura J. Brattain Published: 2026-04-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.27217v1 Cognitive decline is highly heterogeneous across individuals, which complicates prognosis, trial design, and treatment planning. We present the Personalized Cognitive Decline Assessment Digital Twin (PCD-DT), a multimodal and uncertainty-aware framework for modeling patient-specific disease trajectories from sparse, noisy, and irregular longitudinal data. The framework combines three methodological components: (1) latent state-space models for individualized temporal dynamics, (2) multimodal fusion for clinical, biomarker, and imaging features, and (3) uncertainty-aware validation and adaptive updating for robust digital twin operation. We also outline how conditional generative models can support data augmentation and stress testing for underrepresented progression patterns. As a preliminary feasibility study, we analyze longitudinal TADPOLE trajectories and show clear separation between cognitively normal and Alzheimer's disease cohorts in ADAS13, ventricle volume, and hippocampal volume over five years. We further conduct a multimodal next-visit prediction ablation using an LSTM sequence model on 3{,}003 visit-pair sequences derived from TADPOLE, where the combined cognitive plus MRI configuration achieves the lowest standardized RMSE for both ADAS13 (0.4419) and ventricle volume (0.5842), outperforming a Last Observation Carried Forward baseline. A Bayesian tensor modeling component for high-dimensional imaging fusion is also discussed. These results support the feasibility of the proposed architecture while also highlighting the need for stronger uncertainty calibration and longer-horizon predictive evaluation. The PCD-DT framework provides a principled starting point for personalized in silico modeling in neurodegenerative disease. This work positions PCD-DT as a foundational step toward clinically deployable, uncertainty-aware digital twin systems. --- ### Theory Under Construction: Orchestrating Language Models for Research Software Where the Specification Evolves Authors: Halley Young, Nikolaj Björner Published: 2026-04-29 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.27209v2 Large language models can now generate substantial code and draft research text, but research-software projects require more than either artifact alone. The mathematical thesis, executable system, benchmark surface, and public claims must mature together, yet often drift apart. We identify two LM-specific failure modes: hallucination accumulation, in which claims exceed what code or theory supports and unsupported assertions propagate across sessions; and desynchronization, in which code, theory, or the model's own world model fall out of alignment. We propose Comet-H, an iterative prompt automaton that orchestrates ideation, implementation, evaluation, grounding, and paper-writing as coupled coordinates of a single workspace state. At each step, a controller selects the next prompt by scoring it against what the workspace currently lacks, carries unfinished follow-up work forward with a half-life, and re-checks the paper and README against the code and benchmarks whenever documentation changes. We frame prompt selection as a small contextual bandit problem over prompt families, with prompts as arms, workspace deficits as context, and a hand-weighted linear score. This transparent scorer, paired with a fading record of unfinished work, bounds long-horizon follow-ups, requires no learned policy, and makes each prompt choice legible from the workspace. We created a portfolio of 46 research-software repositories across two dozen domains. We study A3 in depth, a Python static-analysis tool built entirely within the loop, which reaches (F1 = 0.768) on a 90-case benchmark, compared with a next-best baseline of 0.364. Across approximately 400 commits, we find that audit-and-contraction passes dominate the later phases of every successful trajectory. --- ### HQ-UNet: A Hybrid Quantum-Classical U-Net with a Quantum Bottleneck for Remote Sensing Image Segmentation Authors: Md Aminur Hossain, Ayush V. Patel, Ikshwaku Vanani, Biplab Banerjee Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.27206v1 Semantic segmentation in remote sensing is commonly addressed using classical deep learning architectures such as U-Net, which require a large number of parameters to model complex spatial relationships. Quantum machine learning (QML) provides an alternative representation paradigm by mapping classical features into quantum states, but its direct application to high-dimensional images remains challenging under near-term quantum hardware constraints. In this work, we propose HQ-UNet, a hybrid quantum-classical U-Net architecture that integrates a compact parameterized quantum circuit at the bottleneck of a classical U-Net. The proposed design uses a non-pooling quantum convolutional module to enrich highly compressed encoder features before decoding, while keeping the quantum component shallow and parameter-efficient. Experiments on the LandCover.ai dataset show that HQ-UNet achieves a mean IoU of 0.8050 and an overall accuracy of 94.76%, outperforming the classical U-Net baseline. These results suggest that compact quantum bottlenecks can enhance feature representation for remote sensing image segmentation under near-term quantum constraints. This highlights the potential of hybrid quantum-classical designs as a promising direction for parameter-efficient dense prediction in Earth observation. ---