# Research Papers — 2026-03 · bundle page 124 of 243 > Full-text markdown of papers #6151–6200 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/124 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/123.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/125.md) ### Point-Identification of a Robust Predictor Under Latent Shift with Imperfect Proxies Authors: Zahra Rahiminasab, Reza Soumi, Arto Klami, Samuel Kaski Published: 2026-03-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.15158v3 Addressing the domain adaptation problem becomes more challenging when distribution shifts across domains stem from latent confounders that affect both covariates and outcomes. Existing proxy-based approaches that address latent shift rely on a strong completeness assumption to uniquely determine (point-identify) a robust predictor. Completeness requires that proxies have sufficient information about variations in latent confounders. For imperfect proxies the mapping from confounders to the space of proxy distributions is non-injective, and multiple latent confounder values can generate the same proxy distribution. This breaks the completeness assumption and observed data are consistent with multiple potential predictors (set-identified). To address this, we introduce latent equivalent classes (LECs). LECs are defined as groups of latent confounders that induce the same conditional proxy distribution. We show that point-identification for the robust predictor remains achievable as long as multiple domains differ sufficiently in how they mix proxy-induced LECs to form the robust predictor. This domain diversity condition is formalized as a cross-domain rank condition on the mixture weights, which is substantially weaker assumption than completeness. We introduce the Proximal Quasi-Bayesian Active learning (PQAL) framework, which actively queries a small, targeted set of diverse domains that satisfy this rank condition. PQAL can recover the point-identified predictor, demonstrates robustness to varying degrees of shift and outperforms previous methods on synthetic data and semi-synthetic dSprites, IHDP, ACS Folktables datasets. --- ### Storage and selection of multiple chaotic attractors in minimal reservoir computers Authors: Francesco Martinuzzi, Holger Kantz Published: 2026-03-16 Categories: nlin.CD, cs.LG arXiv: https://arxiv.org/abs/2603.15155v1 Modern predictive modeling increasingly calls for a single learned dynamical substrate to operate across multiple regimes. From a dynamical-systems viewpoint, this capability decomposes into the storage of multiple attractors and the selection of the appropriate attractor in response to contextual cues. In reservoir computing (RC), multi-attractor learning has largely been pursued using large, randomly wired reservoirs, on the assumption that stochastic connectivity is required to generate sufficiently rich internal dynamics. At the same time, recent work shows that minimal deterministic reservoirs can match random designs for single-system chaotic forecasting. Under which conditions can minimal topologies learn multiple chaotic attractors? In this paper, we find that minimal architectures can successfully store multiple chaotic attractors. However, these same architectures struggle with task switching, in which the system must transition between attractors in response to external cues. We test storage and selection on all 28 unordered system pairs formed from eight three-dimensional chaotic systems. We do not observe a robust dependence of multi-attractor performance on reservoir topology. Over the ten topologies investigated, we find that no single one consistently outperforms the others for either storage or cue-dependent selection. Our results suggest that while minimal substrates possess the representational capacity to model coexisting attractors, they may lack the robust temporal memory required for cued transitions. --- ### Vision-Language Model Based Multi-Expert Fusion for CT Image Classification Authors: Jianfa Bai, Kejin Lu, Runtian Yuan, Qingqiu Li, Jilan Xu, Junlin Hou, Yuejie Zhang, Rui Feng Published: 2026-03-16 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2603.15154v1 Robust detection of COVID-19 from chest CT remains challenging in multi-institutional settings due to substantial source shift, source imbalance, and hidden test-source identities. In this work, we propose a three-stage source-aware multi-expert framework for multi-source COVID-19 CT classification. First, we build a lung-aware 3D expert by combining original CT volumes and lung-extracted CT volumes for volumetric classification. Second, we develop two MedSigLIP-based experts: a slice-wise representation and probability learning module, and a Transformer-based inter-slice context modeling module for capturing cross-slice dependency. Third, we train a source classifier to predict the latent source identity of each test scan. By leveraging the predicted source information, we perform model fusion and voting based on different experts. On the validation set covering all four sources, the Stage 1 model achieves the best macro-F1 of 0.9711, ACC of 0.9712, and AUC of 0.9791. Stage~2a and Stage~2b achieve the best AUC scores of 0.9864 and 0.9854, respectively. Stage~3 source classifier reaches 0.9107 ACC and 0.9114 F1. These results demonstrate that source-aware expert modeling and hierarchical voting provide an effective solution for robust COVID-19 CT classification under heterogeneous multi-source conditions. --- ### TextOVSR: Text-Guided Real-World Opera Video Super-Resolution Authors: Hua Chang, Xin Xu, Wei Liu, Jiayi Wu, Kui Jiang, Fei Ma, Qi Tian Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15153v1 Many classic opera videos exhibit poor visual quality due to the limitations of early filming equipment and long-term degradation during storage. Although real-world video super-resolution (RWVSR) has achieved significant advances in recent years, directly applying existing methods to degraded opera videos remains challenging. The difficulties are twofold. First, accurately modeling real-world degradations is complex: simplistic combinations of classical degradation kernels fail to capture the authentic noise distribution, while methods that extract real noise patches from external datasets are prone to style mismatches that introduce visual artifacts. Second, current RWVSR methods, which rely solely on degraded image features, struggle to reconstruct realistic and detailed textures due to a lack of high-level semantic guidance. To address these issues, we propose a Text-guided Dual-Branch Opera Video Super-Resolution (TextOVSR) network, which introduces two types of textual prompts to guide the super-resolution process. Specifically, degradation-descriptive text, derived from the degradation process, is incorporated into the negative branch to constrain the solution space. Simultaneously, content-descriptive text is incorporated into a positive branch and our proposed Text-Enhanced Discriminator (TED) to provide semantic guidance for enhanced texture reconstruction. Furthermore, we design a Degradation-Robust Feature Fusion (DRF) module to facilitate cross-modal feature fusion while suppressing degradation interference. Experiments on our OperaLQ benchmark show that TextOVSR outperforms state-of-the-art methods both qualitatively and quantitatively. The code is available at https://github.com/ChangHua0/TextOVSR. --- ### SNCE: Geometry-Aware Supervision for Scalable Discrete Image Generation Authors: Shufan Li, Jiuxiang Gu, Kangning Liu, Zhe Lin, Aditya Grover, Jason Kuen Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15150v1 Recent advancements in discrete image generation showed that scaling the VQ codebook size significantly improves reconstruction fidelity. However, training generative models with a large VQ codebook remains challenging, typically requiring larger model size and a longer training schedule. In this work, we propose Stochastic Neighbor Cross Entropy Minimization (SNCE), a novel training objective designed to address the optimization challenges of large-codebook discrete image generators. Instead of supervising the model with a hard one-hot target, SNCE constructs a soft categorical distribution over a set of neighboring tokens. The probability assigned to each token is proportional to the proximity between its code embedding and the ground-truth image embedding, encouraging the model to capture semantically meaningful geometric structure in the quantized embedding space. We conduct extensive experiments across class-conditional ImageNet-256 generation, large-scale text-to-image synthesis, and image editing tasks. Results show that SNCE significantly improves convergence speed and overall generation quality compared to standard cross-entropy objectives. --- ### Accelerating Byzantine-Robust Distributed Learning with Compressed Communication via Double Momentum and Variance Reduction Authors: Yanghao Li, Changxin Liu, Yuhao Yi Published: 2026-03-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.15144v2 In collaborative and distributed learning, Byzantine robustness reflects a major facet of optimization algorithms. Such distributed algorithms are often accompanied by transmitting a large number of parameters, so communication compression is essential for an effective solution. In this paper, we propose Byz-DM21, a novel Byzantine-robust and communication-efficient stochastic distributed learning algorithm. Our key innovation is a novel gradient estimator based on a double-momentum mechanism, integrating recent advancements in error feedback techniques. Using this estimator, we design both standard and accelerated algorithms that eliminate the need for large batch sizes while maintaining robustness against Byzantine workers. We prove that the Byz-DM21 algorithm has a smaller neighborhood size and converges to $\varepsilon$-stationary points in $\mathcal{O}(\varepsilon^{-4})$ iterations. To further enhance efficiency, we introduce a distributed variant called Byz-VR-DM21, which incorporates local variance reduction at each node to progressively eliminate variance from random approximations. We show that Byz-VR-DM21 provably converges to $\varepsilon$-stationary points in $\mathcal{O}(\varepsilon^{-3 })$ iterations. Additionally, we extend our results to the case where the functions satisfy the Polyak-Łojasiewicz condition. Finally, numerical experiments demonstrate the effectiveness of the proposed method. --- ### Clinical Priors Guided Lung Disease Detection in 3D CT Scans Authors: Kejin Lu, Jianfa Bai, Qingqiu Li, Runtian Yuan, Jilan Xu, Junlin Hou, Yuejie Zhang, Rui Feng Published: 2026-03-16 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2603.15143v2 Accurate classification of lung diseases from chest CT scans plays an important role in computer-aided diagnosis systems. However, medical imaging datasets often suffer from severe class imbalance, which may significantly degrade the performance of deep learning models, especially for minority disease categories. To address this issue, we propose a gender-aware two-stage lung disease classification framework. The proposed approach explicitly incorporates gender information into the disease recognition pipeline. In the first stage, a gender classifier is trained to predict the patient's gender from CT scans. In the second stage, the input CT image is routed to a corresponding gender-specific disease classifier to perform final disease prediction. This design enables the model to better capture gender-related imaging characteristics and alleviate the influence of imbalanced data distribution. Experimental results demonstrate that the proposed method improves the recognition performance for minority disease categories, particularly squamous cell carcinoma, while maintaining competitive performance on other classes. --- ### Context-Aware Sensor Modeling for Asynchronous Multi-Sensor Tracking in Stone Soup Authors: Martin Vonheim Larsen, Kim Mathiassen Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15137v1 Multi-sensor tracking in the real world involves asynchronous sensors with partial coverage and heterogeneous detection performance. Although probabilistic tracking methods permit detection probability and clutter intensity to depend on state and sensing context, many practical frameworks enforce globally uniform observability assumptions. Under multi-rate and partially overlapping sensing, this simplification causes repeated non-detections from high-rate sensors to erode tracks visible only to low-rate sensors, potentially degrading fusion performance. We introduce DetectorContext, an abstraction for the open-source multi-target tracking framework Stone Soup. DetectorContext exposes detection probability and clutter intensity as state-dependent functions evaluated during hypothesis formation. The abstraction integrates with existing probabilistic trackers without modifying their update equations. Experiments on asynchronous radar-lidar data demonstrate that context-aware modeling restores stable fusion and significantly improves HOTA and GOSPA performance without increasing false tracks. --- ### Safe Flow Q-Learning: Offline Safe Reinforcement Learning with Reachability-Based Flow Policies Authors: Mumuksh Tayal, Manan Tayal, Ravi Prakash Published: 2026-03-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.15136v2 Offline safe reinforcement learning (RL) seeks reward-maximizing policies from static datasets under strict safety constraints. Existing methods often rely on soft expected-cost objectives or iterative generative inference, which can be insufficient for safety-critical real-time control. We propose Safe Flow Q-Learning (SafeFQL), which extends FQL to safe offline RL by combining a Hamilton--Jacobi reachability-inspired safety value function with an efficient one-step flow policy. SafeFQL learns the safety value via a self-consistency Bellman recursion, trains a flow policy by behavioral cloning, and distills it into a one-step actor for reward-maximizing safe action selection without rejection sampling at deployment. Empirically, SafeFQL trades modestly higher offline training cost for substantially lower inference latency than diffusion-style safe generative baselines, which is advantageous for real-time safety-critical deployment. Across boat navigation, and Safety Gymnasium MuJoCo tasks, SafeFQL matches or exceeds prior offline safe RL performance while substantially reducing constraint violations. --- ### WiT: Waypoint Diffusion Transformers for Alleviating Trajectory Conflict in Pixel-Space Image Generation Authors: Hainuo Wang, Mingjia Li, Xiaojie Guo Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15132v3 While recent Flow Matching models avoid the reconstruction bottlenecks of latent autoencoders by operating directly in pixel space, the raw pixel manifold provides little explicit semantic organization, making target-specific transport directions difficult for a shared finite-capacity vector field to distinguish. When different transport requirements are locally entangled in this way, their learning signals can interfere and hinder optimization, a phenomenon we refer to as trajectory conflict. To alleviate trajectory conflict while preserving direct generation in pixel space, we propose Waypoint Diffusion Transformers (WiT), which introduces explicit semantic routing into pixel-space generation. WiT structures pixel-space prediction through compact semantic waypoints projected from pre-trained vision representations, providing a semantically organized intermediate representation that complements direct pixel prediction. During ODE integration, a lightweight Waypoints Predictor dynamically infers the semantic waypoint from the current noisy state, and the predicted waypoint continuously conditions the primary Pixel Space Generator through our Just-Pixel AdaLN mechanism. This spatially varying semantic modulation reorganizes the finite-capacity learning problem and helps the model preserve target-specific transport directions while retaining generation entirely in pixel space. Experiments on ImageNet 2012 demonstrate consistent improvements over strong pixel-space baselines across model scales, with WiT-H/16 achieving an FID of 1.79. Code will be released at https://hainuo-wang.github.io/WiT. --- ### WiT: Waypoint Diffusion Transformers via Trajectory Conflict Navigation Authors: Hainuo Wang, Mingjia Li, Xiaojie Guo Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15132v2 While recent Flow Matching models avoid the reconstruction bottlenecks of latent autoencoders by operating directly in pixel space, the lack of semantic continuity in the pixel manifold severely intertwines optimal transport paths. This induces severe trajectory conflicts near intersections, yielding sub-optimal solutions. Rather than bypassing this issue via information-lossy latent representations, we directly untangle the pixel-space trajectories by proposing Waypoint Diffusion Transformers (WiT). WiT factorizes the continuous vector field via intermediate semantic waypoints projected from pre-trained vision models. It effectively disentangles the generation trajectories by breaking the optimal transport into prior-to-waypoint and waypoint-to-pixel segments. Specifically, during the iterative denoising process, a lightweight generator dynamically infers these intermediate waypoints from the current noisy state. They then continuously condition the primary diffusion transformer via the Just-Pixel AdaLN mechanism, steering the evolution towards the next state, ultimately yielding the final RGB pixels. Evaluated on ImageNet 256x256, WiT beats strong pixel-space baselines, accelerating JiT training convergence by 2.2x. Code will be publicly released at https://github.com/hainuo-wang/WiT.git. --- ### Low-light Image Enhancement with Retinex Decomposition in Latent Space Authors: Bolun Zheng, Qingshan Lei, Quan Chen, Qianyu Zhang, Kainan Yu, Xu Jia, Lingyu Zhu Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15131v1 Retinex theory provides a principled foundation for low-light image enhancement, inspiring numerous learning-based methods that integrate its principles. However, existing methods exhibits limitations in accurately decomposing reflectance and illumination components. To address this, we propose a Retinex-Guided Transformer~(RGT) model, which is a two-stage model consisting of decomposition and enhancement phases. First, we propose a latent space decomposition strategy to separate reflectance and illumination components. By incorporating the log transformation and 1-pixel offset, we convert the intrinsically multiplicative relationship into an additive formulation, enhancing decomposition stability and precision. Subsequently, we construct a U-shaped component refiner incorporating the proposed guidance fusion transformer block. The component refiner refines reflectance component to preserve texture details and optimize illumination distribution, effectively transforming low-light inputs to normal-light counterparts. Experimental evaluations across four benchmark datasets validate that our method achieves competitive performance in low-light enhancement and a more stable training process. --- ### Indirect Question Answering in English, German and Bavarian: A Challenging Task for High- and Low-Resource Languages Alike Authors: Miriam Winkler, Verena Blaschke, Barbara Plank Published: 2026-03-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.15130v3 Indirectness is a common feature of daily communication, yet is underexplored in NLP research for both low-resource as well as high-resource languages. Indirect Question Answering (IQA) aims at classifying the polarity of indirect answers. In this paper, we present two multilingual corpora for IQA of varying quality that both cover English, Standard German and Bavarian, a German dialect without standard orthography: InQA+, a small high-quality evaluation dataset with hand-annotated labels, and GenIQA, a larger training dataset, that contains artificial data generated by GPT-4o-mini. We find that IQA is a pragmatically hard task that comes with various challenges, based on several experiment variations with multilingual transformer models (mBERT, XLM-R and mDeBERTa). We suggest and employ recommendations to tackle these challenges. Our results reveal low performance, even for English, and severe overfitting. We analyse various factors that influence these results, including label ambiguity, label set and dataset size. We find that the IQA performance is poor in high- (English, German) and low-resource languages (Bavarian) and that it is beneficial to have a large amount of training data. Further, GPT-4o-mini does not possess enough pragmatic understanding to generate high-quality IQA data in any of our tested languages. --- ### Next-Frame Decoding for Ultra-Low-Bitrate Image Compression with Video Diffusion Priors Authors: Yunuo Chen, Chuqin Zhou, Jiangchuan Li, Xiaoyue Ling, Bing He, Jincheng Dai, Li Song, Guo Lu Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15129v3 We present a novel paradigm for ultra-low-bitrate image compression (ULB-IC) that exploits the ``temporal'' evolution in generative image compression. Specifically, we define an explicit intermediate state during decoding: a compact anchor frame, which preserves the scene geometry and semantic layout while discarding high-frequency details. We then reinterpret generative decoding as a virtual temporal transition from this anchor to the final reconstructed image. To model this progression, we leverage a pretrained video diffusion model (VDM) as a temporal prior: the anchor frame serves as the initial frame and the original image as the target frame, transforming the decoding process into a next-frame prediction task. In contrast to image diffusion-based ULB-IC models, our decoding proceeds from a visible, semantically faithful anchor, which improves both fidelity and realism for perceptual image compression. Extensive experiments demonstrate that our method achieves superior rate-distortion performance. On the CLIC2020 test set, our method achieves over 50% bitrate savings across LPIPS, DISTS, FID, and KID compared to DiffC, while also delivering a significant decoding speedup of up to $\times$5. Code will be released at https://github.com/UnoC-727/NeFIC. --- ### A Novel Camera-to-Robot Calibration Method for Vision-Based Floor Measurements Authors: Jan Andre Rudolph, Dennis Haitz, Markus Ulrich Published: 2026-03-16 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2603.15126v2 A novel hand-eye calibration method for ground-observing mobile robots is proposed. While cameras on mobile robots are common, they are rarely used for ground-observing measurement tasks. Laser trackers are increasingly used in robotics for precise localization. A referencing plate is designed to combine the two measurement modalities of laser-tracker 3D metrology and camera-based 2D imaging. It incorporates reflector nests for pose acquisition using a laser tracker and a camera calibration target that is observed by the robot-mounted camera. The procedure comprises estimating the plate pose, the plate-camera pose, and the robot pose, followed by computing the robot-camera transformation. Experiments indicate sub-millimeter repeatability. --- ### Establishing Construct Validity in LLM Capability Benchmarks Requires Nomological Networks Authors: Timo Freiesleben Published: 2026-03-16 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.15121v1 Recent work in machine learning increasingly attributes human-like capabilities such as reasoning or theory of mind to large language models (LLMs) on the basis of benchmark performance. This paper examines this practice through the lens of construct validity, understood as the problem of linking theoretical capabilities to their empirical measurements. It contrasts three influential frameworks: the nomological account developed by Cronbach and Meehl, the inferential account proposed by Messick and refined by Kane, and Borsboom's causal account. I argue that the nomological account provides the most suitable foundation for current LLM capability research. It avoids the strong ontological commitments of the causal account while offering a more substantive framework for articulating construct meaning than the inferential account. I explore the conceptual implications of adopting the nomological account for LLM research through a concrete case: the assessment of reasoning capabilities in LLMs. --- ### A Tutorial on ALOS2 SAR Utilization: Dataset Preparation, Self-Supervised Pretraining, and Semantic Segmentation Authors: Nevrez Imamoglu, Ali Caglayan, Toru Kouyama Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15119v2 Masked auto-encoders (MAE) and related approaches have shown promise for satellite imagery, but their application to synthetic aperture radar (SAR) remains limited due to challenges in semantic labeling and high noise levels. Building on our prior work with SAR-W-MixMAE, which adds SAR-specific intensity-weighted loss to standard MixMAE for pretraining, we also introduce SAR-W-SimMIM; a weighted variant of SimMIM applied to ALOS-2 single-channel SAR imagery. This method aims to reduce the impact of speckle and extreme intensity values during self-supervised pretraining. We evaluate its effect on semantic segmentation compared to our previous trial with SAR-W-MixMAE and random initialization, observing notable improvements. In addition, pretraining and fine-tuning models on satellite imagery pose unique challenges, particularly when developing region-specific models. Imbalanced land cover distributions such as dominant water, forest, or desert areas can introduce bias, affecting both pretraining and downstream tasks like land cover segmentation. To address this, we constructed a SAR dataset using ALOS-2 single-channel (HH polarization) imagery focused on the Japan region, marking the initial phase toward a national-scale foundation model. This dataset was used to pretrain a vision transformer-based autoencoder, with the resulting encoder fine-tuned for semantic segmentation using a task-specific decoder. Initial results demonstrate significant performance improvements compared to training from scratch with random initialization. In summary, this work provides a guide to process and prepare ALOS2 observations to create dataset so that it can be taken advantage of self-supervised pretraining of models and finetuning downstream tasks such as semantic segmentation. --- ### VAREX: A Benchmark for Multi-Modal Structured Extraction from Documents Authors: Udi Barzelay, Ophir Azulai, Inbar Shapira, Idan Friedman, Foad Abo Dahood, Madison Lee, Abraham Daniels Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15118v2 We introduce VAREX (VARied-schema EXtraction), a benchmark for evaluating multimodal foundation models on structured data extraction from government forms. VAREX employs a Reverse Annotation pipeline that programmatically fills PDF templates with synthetic values, producing deterministic ground truth validated through three-phase quality assurance. The benchmark comprises 1,777 documents with 1,771 unique schemas across three structural categories, each provided in four input modalities: plain text, layout-preserving text (whitespace-aligned to approximate column positions), document image, or both text and image combined. Unlike existing benchmarks that evaluate from a single input representation, VAREX provides four controlled modalities per document, enabling systematic ablation of how input format affects extraction accuracy -- a capability absent from prior benchmarks. We evaluate 20 models from frontier proprietary models to small open models, with particular attention to models <=4B parameters suitable for cost-sensitive and latency-constrained deployment. Results reveal that (1) below 4B parameters, structured output compliance -- not extraction capability -- is a dominant bottleneck; in particular, schema echo (models producing schema-conforming structure instead of extracted values) depresses scores by 45-65 pp (percentage points) in affected models; (2) extraction-specific fine-tuning at 2B yields +81 pp gains, demonstrating that the instruction-following deficit is addressable without scale; (3) layout-preserving text provides the largest accuracy gain (+3-18 pp), exceeding pixel-level visual cues; and (4) the benchmark most effectively discriminates models in the 60-95% accuracy band. Dataset and evaluation code are publicly available. --- ### MMKU-Bench: A Multimodal Update Benchmark for Diverse Visual Knowledge Authors: Baochen Fu, Yuntao Du, Cheng Chang, Baihao Jin, Wenzhi Deng, Muhao Xu, Hongmei Yan, Weiye Song, Yi Wan Published: 2026-03-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.15117v1 As real-world knowledge continues to evolve, the parametric knowledge acquired by multimodal models during pretraining becomes increasingly difficult to remain consistent with real-world knowledge. Existing research on multimodal knowledge updating focuses only on learning previously unknown knowledge, while overlooking the need to update knowledge that the model has already mastered but that later changes; moreover, evaluation is limited to the same modality, lacking a systematic analysis of cross-modal consistency. To address these issues, this paper proposes MMKU-Bench, a comprehensive evaluation benchmark for multimodal knowledge updating, which contains over 25k knowledge instances and more than 49k images, covering two scenarios, updated knowledge and unknown knowledge, thereby enabling comparative analysis of learning across different knowledge types. On this benchmark, we evaluate a variety of representative approaches, including supervised fine-tuning (SFT), reinforcement learning from human feedback (RLHF), and knowledge editing (KE). Experimental results show that SFT and RLHF are prone to catastrophic forgetting, while KE better preserve general capabilities but exhibit clear limitations in continual updating. Overall, MMKU-Bench provides a reliable and comprehensive evaluation benchmark for multimodal knowledge updating, advancing progress in this field. --- ### Sampling-guided exploration of active feature selection policies Authors: Gabriel Bernardino, Anders Jonsson, Patrick Clarysse, Nicolas Duchateau Published: 2026-03-16 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2603.15110v1 Determining the most appropriate features for machine learning predictive models is challenging regarding performance and feature acquisition costs. In particular, global feature choice is limited given that some features will only benefit a subset of instances. In previous work, we proposed a reinforcement learning approach to sequentially recommend which modality to acquire next to reach the best information/cost ratio, based on the instance-specific information already acquired. We formulated the problem as a Markov Decision Process where the state's dimensionality changes during the episode, avoiding data imputation, contrary to existing works. However, this only allowed processing a small number of features, as all possible combinations of features were considered. Here, we address these limitations with two contributions: 1) we expand our framework to larger datasets with a heuristic-based strategy that focuses on the most promising feature combinations, and 2) we introduce a post-fit regularisation strategy that reduces the number of different feature combinations, leading to compact sequences of decisions. We tested our method on four binary classification datasets (one involving high-dimensional variables), the largest of which had 56 features and 4500 samples. We obtained better performance than state-of-the-art methods, both in terms of accuracy and policy complexity. --- ### PAKAN: Pixel Adaptive Kolmogorov-Arnold Network Modules for Pansharpening Authors: Haoyu Zhang, Haojing Chen, Zhen Zhong, Liangjian Deng Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15109v1 Pansharpening aims to fuse high-resolution spatial details from panchromatic images with the rich spectral information of multispectral images. Existing deep neural networks for this task typically rely on static activation functions, which limit their ability to dynamically model the complex, non-linear mappings required for optimal spatial-spectral fusion. While the recently introduced Kolmogorov-Arnold Network (KAN) utilizes learnable activation functions, traditional KANs lack dynamic adaptability during inference. To address this limitation, we propose a Pixel Adaptive Kolmogorov-Arnold Network framework. Starting from KAN, we design two adaptive variants: a 2D Adaptive KAN that generates spline summation weights across spatial dimensions and a 1D Adaptive KAN that generates them across spectral channels. These two components are then assembled into PAKAN 2to1 for feature fusion and PAKAN 1to1 for feature refinement. Extensive experiments demonstrate that our proposed modules significantly enhance network performance, proving the effectiveness and superiority of pixel-adaptive activation in pansharpening tasks. --- ### PrototypeNAS: Rapid Design of Deep Neural Networks for Microcontroller Units Authors: Mark Deutel, Simon Geis, Axel Plinge Published: 2026-03-16 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.15106v2 Enabling efficient deep neural network (DNN) inference on edge devices with different hardware constraints is a challenging task that typically requires DNN architectures to be specialized for each device separately. To avoid the huge manual effort, one can use neural architecture search (NAS). However, many existing NAS methods are resource-intensive and time-consuming because they require the training of many different DNNs from scratch. Furthermore, they do not take the resource constraints of the target system into account. To address these shortcomings, we propose PrototypeNAS, a zero-shot NAS method to accelerate and automate the selection, compression, and specialization of DNNs to different target microcontroller units (MCUs). We propose a novel three-step search method that decouples DNN design and specialization from DNN training for a given target platform. First, we present a novel search space that not only cuts out smaller DNNs from a single large architecture, but instead combines the structural optimization of multiple architecture types, as well as optimization of their pruning and quantization configurations. Second, we explore the use of an ensemble of zero-shot proxies during optimization instead of a single one. Third, we propose the use of Hypervolume subset selection to distill DNN architectures from the Pareto front of the multi-objective optimization that represent the most meaningful tradeoffs between accuracy and FLOPs. We evaluate the effectiveness of PrototypeNAS on 12 different datasets in three different tasks: image classification, time series classification, and object detection. Our results demonstrate that PrototypeNAS is able to identify DNN models within minutes that are small enough to be deployed on off-the-shelf MCUs and still achieve accuracies comparable to the performance of large DNN models. --- ### Learning from Limited and Incomplete Data: A Multimodal Framework for Predicting Pathological Response in NSCLC Authors: Alice Natalina Caragliano, Giulia Farina, Fatih Aksu, Camillo Maria Caruso, Claudia Tacconi, Carlo Greco, Lorenzo Nibid, Edy Ippolito, Michele Fiore, Giuseppe Perrone, Sara Ramella, Paolo Soda et al. Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15100v1 Major pathological response (pR) following neoadjuvant therapy is a clinically meaningful endpoint in non-small cell lung cancer, strongly associated with improved survival. However, accurate preoperative prediction of pR remains challenging, particularly in real-world clinical settings characterized by limited data availability and incomplete clinical profiles. In this study, we propose a multimodal deep learning framework designed to address these constraints by integrating foundation model-based CT feature extraction with a missing-aware architecture for clinical variables. This approach enables robust learning from small cohorts while explicitly modeling missing clinical information, without relying on conventional imputation strategies. A weighted fusion mechanism is employed to leverage the complementary contributions of imaging and clinical modalities, yielding a multimodal model that consistently outperforms both unimodal imaging and clinical baselines. These findings underscore the added value of integrating heterogeneous data sources and highlight the potential of multimodal, missing-aware systems to support pR prediction under realistic clinical conditions. --- ### Bridging National and International Legal Data: Two Projects Based on the Japanese Legal Standard XML Schema for Comparative Law Studies Authors: Makoto Nakamura Published: 2026-03-16 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.15094v1 This paper presents an integrated framework for computational comparative law by connecting two consecutive research projects based on the Japanese Legal Standard (JLS) XML schema. The first project establishes structural interoperability by developing a conversion pipeline from JLS to the Akoma Ntoso (AKN) standard, enabling Japanese statutes to be integrated into international LegalDocML-based legislative databases. Building on this foundation, the second project applies multilingual embedding models and semantic textual similarity techniques to identify corresponding provisions across national legal systems. A prototype system combining multilingual embeddings, FAISS retrieval, and Cross-Encoder reranking generates candidate correspondences and visualizes them as cross-jurisdictional networks for exploratory comparative analysis. --- ### Trustworthy Koopman Operator Learning: Invariance Diagnostics and Error Bounds Authors: Gustav Conradie, Nicolas Boullé, Jean-Christophe Loiseau, Steven L. Brunton, Matthew J. Colbrook Published: 2026-03-16 Categories: math.NA, cs.LG, math.DS, math.OC arXiv: https://arxiv.org/abs/2603.15091v1 Koopman operator theory provides a global linear representation of nonlinear dynamics and underpins many data-driven methods. In practice, however, finite-dimensional feature spaces induced by a user-chosen dictionary are rarely invariant, so closure failures and projection errors lead to spurious eigenvalues, misleading Koopman modes, and overconfident forecasts. This paper addresses a central validation problem in data-driven Koopman methods: how to quantify invariance and projection errors for an arbitrary feature space using only snapshot data, and how to use these diagnostics to produce actionable guarantees and guide dictionary refinement? A unified a posteriori methodology is developed for certifying when a Koopman approximation is trustworthy and improving it when it is not. Koopman invariance is quantified using principal angles between a subspace and its Koopman image, yielding principal observables and a principal angle decomposition (PAD), a dynamics-informed alternative to SVD truncation with significantly improved performance. Multi-step error bounds are derived for Koopman and Perron--Frobenius mode decompositions, including RKHS-based pointwise guarantees, and are complemented by Gaussian process expected error surrogates. The resulting toolbox enables validated spectral analysis, certified forecasting, and principled dictionary and kernel learning, demonstrated on chaotic and high-dimensional benchmarks and real-world datasets, including cavity flow and the Pluto--Charon system. --- ### Affordable Precision Agriculture: A Deployment-Oriented Review of Low-Cost, Low-Power Edge AI and TinyML for Resource-Constrained Farming Systems Authors: Riya Samanta, Bidyut Saha Published: 2026-03-16 Categories: cs.ET, cs.LG arXiv: https://arxiv.org/abs/2603.15085v1 Precision agriculture increasingly integrates artificial intelligence to enhance crop monitoring, irrigation management, and resource efficiency. Nevertheless, the vast majority of the current systems are still mostly cloud-based and require reliable connectivity, which hampers the adoption to smaller scale, smallholder farming and underdeveloped country systems. Using recent literature reviews, ranging from 2023 to 2026, this review covers deployments of Edge AI, focused on the evolution and acceptance of Tiny Machine Learning, in low-cost and low-powered agriculture. A hardware-targeted deployment-oriented study has shown pronounced variation in architecture with microcontroller-class platforms i.e. ESP32, STM32, ATMega dominating the inference options, in parallel with single-board computers and UAV-assisted solutions. Quantitative synthesis shows quantization is the dominant optimization strategy; the approach in many works identified: around 50% of such works are quantized, while structured pruning, multi-objective compression and hardware aware neural architecture search are relatively under-researched. Also, resource profiling practices are not uniform: while model size is occasionally reported, explicit flash, RAM, MAC, latency and millijoule level energy metrics are not well documented, hampering reproducibility and cross-system comparison. Moreoever, to bridge the gap between research prototypes and deployment-ready systems, the review also presents a literature-informed deployment perspective in the form of a privacy-preserving layered Edge AI architecture for agriculture, synthesizing the key system-level design insights emerging from the surveyed works. Overall, the findings demonstrate a clear architectural shift toward localized inference with centralized training asymmetry. --- ### ReactMotion: Generating Reactive Listener Motions from Speaker Utterance Authors: Cheng Luo, Bizhu Wu, Bing Li, Jianfeng Ren, Ruibin Bai, Rong Qu, Linlin Shen, Bernard Ghanem Published: 2026-03-16 Categories: cs.CV, cs.AI, cs.HC, cs.MM, cs.SD arXiv: https://arxiv.org/abs/2603.15083v1 In this paper, we introduce a new task, Reactive Listener Motion Generation from Speaker Utterance, which aims to generate naturalistic listener body motions that appropriately respond to a speaker's utterance. However, modeling such nonverbal listener behaviors remains underexplored and challenging due to the inherently non-deterministic nature of human reactions. To facilitate this task, we present ReactMotionNet, a large-scale dataset that pairs speaker utterances with multiple candidate listener motions annotated with varying degrees of appropriateness. This dataset design explicitly captures the one-to-many nature of listener behavior and provides supervision beyond a single ground-truth motion. Building on this dataset design, we develop preference-oriented evaluation protocols tailored to evaluate reactive appropriateness, where conventional motion metrics focusing on input-motion alignment ignore. We further propose ReactMotion, a unified generative framework that jointly models text, audio, emotion, and motion, and is trained with preference-based objectives to encourage both appropriate and diverse listener responses. Extensive experiments show that ReactMotion outperforms retrieval baselines and cascaded LLM-based pipelines, generating more natural, diverse, and appropriate listener motions. --- ### Open Biomedical Knowledge Graphs at Scale: Construction, Federation, and AI Agent Access with Samyama Graph Database Authors: Madhulatha Mandarapu, Sandeep Kunkunuru Published: 2026-03-16 Categories: cs.DB, cs.AI, q-bio.QM arXiv: https://arxiv.org/abs/2603.15080v3 Biomedical knowledge is fragmented across siloed databases -- Reactome for pathways, STRING for protein interactions, ClinicalTrials.gov for study registries, DrugBank for drug vocabularies, DGIdb for drug-gene interactions, SIDER for side effects. We present three open-source biomedical knowledge graphs -- Pathways KG (118,686 nodes, 834,785 edges from 5 sources), Clinical Trials KG (7,774,446 nodes, 26,973,997 edges from 5 sources), and Drug Interactions KG (32,726 nodes, 191,970 edges from 3 sources) -- built on Samyama, a high-performance graph database written in Rust. Our contributions are threefold. First, we describe a reproducible ETL pattern for constructing large-scale KGs from heterogeneous public data sources, with cross-source deduplication, batch loading (Python Cypher and Rust native loaders), and portable snapshot export. Second, we demonstrate cross-KG federation: loading all three snapshots into a single graph tenant enables property-based joins across datasets. Third, we introduce schema-driven MCP server generation for LLM agent access, evaluated on a new BiomedQA benchmark (40 pharmacology questions): domain-specific MCP tools achieve 98% accuracy vs. 85% for schema-aware text-to-Cypher and 75% for standalone GPT-4o, with zero schema errors. All data sources are open-license. The combined federated graph (7.9M nodes, 28M edges) loads in approximately 3 minutes on commodity cloud hardware, with single-KG queries completing in 80-100ms and cross-KG federation joins in 1-4s --- ### Interpretable Classification of Time Series Using Euler Characteristic Surfaces Authors: Salam Rabindrajit Luwang, Sushovan Majhi, Vishal Mandal, Atish J. Mitra, Md. Nurujjaman, Buddha Nath Sharma Published: 2026-03-16 Categories: cs.LG, math.AT arXiv: https://arxiv.org/abs/2603.15079v1 Persistent homology (PH) -- the conventional method in topological data analysis -- is computationally expensive, requires further vectorization of its signatures before machine learning (ML) can be applied, and captures information along only the spatial axis. For time series data, we propose Euler Characteristic Surfaces (ECS) as an alternative topological signature based on the Euler characteristic ($χ$) -- a fundamental topological invariant. The ECS provides a computationally efficient, spatiotemporal, and inherently discretized feature representation that can serve as direct input to ML models. We prove a stability theorem guaranteeing that the ECS remains stable under small perturbations of the input time series. We first demonstrate that ECS effectively captures the nontrivial topological differences between the limit cycle and the strange attractor in the Rössler system. We then develop an ECS-based classification framework and apply it to five benchmark biomedical datasets (four ECG, one EEG) from the UCR/UEA archive. On $\textit{ECG5000}$, our single-feature ECS classifier achieves $98\%$ accuracy with $O(n+R\cdot T)$ complexity, compared to $62\%$ reported by a recent PH-based method. An AdaBoost extension raises accuracy to $98.6\%$, matching the best deep learning results while retaining full interpretability. Strong results are also obtained on $\textit{TwoLeadECG}$ ($94.1\%$) and $\textit{Epilepsy2}$ ($92.6\%$). --- ### Generative Semantic HARQ: Latent-Space Text Retransmission and Combining Authors: Bin Han, Yulin Hu, Hans D. Schotten Published: 2026-03-16 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2603.15068v1 Semantic communication conveys meaning rather than raw bits, but reliability at the semantic level remains an open challenge. We propose a semantic-level hybrid automatic repeat request (HARQ) framework for text communication, in which a Transformer-variational autoencoder (VAE) codec operates as a lightweight overlay on the conventional protocol stack. The stochastic encoder inherently generates diverse latent representations across retransmissions-providing incremental knowledge (IK) from a single model without dedicated protocol design. On the receiver side, a soft quality estimator triggers retransmissions and a quality-aware combiner merges the received latent vectors within a consistent latent space. We systematically benchmark six semantic quality metrics and four soft combining strategies under hybrid semantic distortion that mixes systematic bias with additive noise. The results suggest combining Weighted-Average or MRC-Inspired combining with self-consistency-based HARQ triggering for the best performance. --- ### The Good, the Better, and the Best: Improving the Discriminability of Face Embeddings through Attribute-aware Learning Authors: Ana Dias, João Ribeiro Pinto, Hugo Proença, João C. Neves Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15062v1 Despite recent advances in face recognition, robust performance remains challenging under large variations in age, pose, and occlusion. A common strategy to address these issues is to guide representation learning with auxiliary supervision from facial attributes, encouraging the visual encoder to focus on identity-relevant regions. However, existing approaches typically rely on heterogeneous and fixed sets of attributes, implicitly assuming equal relevance across attributes. This assumption is suboptimal, as different attributes exhibit varying discriminative power for identity recognition, and some may even introduce harmful biases. In this paper, we propose an attribute-aware face recognition architecture that supervises the learning of facial embeddings using identity class labels, identity-relevant facial attributes, and non-identity-related attributes. Facial attributes are organized into interpretable groups, making it possible to decompose and analyze their individual contributions in a human-understandable manner. Experiments on standard face verification benchmarks demonstrate that joint learning of identity and facial attributes improves the discriminability of face embeddings with two major conclusions: (i) using identity-relevant subsets of facial attributes consistently outperforms supervision with a broader attribute set, and (ii) explicitly forcing embeddings to unlearn non-identity-related attributes yields further performance gains compared to leaving such attributes unsupervised. Additionally, our method serves as a diagnostic tool for assessing the trustworthiness of face recognition encoders by allowing for the measurement of accuracy gains with suppression of non-identity-relevant attributes, with such gains suggesting shortcut learning from redundant attributes associated with each identity. --- ### Writer-R1: Enhancing Generative Writing in LLMs via Memory-augmented Replay Policy Optimization Authors: Jihao Zhao, Shuaishuai Zu, Zhiyuan Ji, Chunlai Zhou, Biao Qin Published: 2026-03-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.15061v1 As a typical open-ended generation task, creative writing lacks verifiable reference answers, which has long constrained reward modeling and automatic evaluation due to high human annotation costs, evaluative bias, and coarse feedback signals. To address these challenges, this paper first designs a multi-agent collaborative workflow based on Grounded Theory, performing dimensional decomposition and hierarchical induction of the problem to dynamically produce interpretable and reusable fine-grained criteria. Furthermore, we propose the Memory-augmented Replay Policy Optimization (MRPO) algorithm: on the one hand, without additional training, MRPO guides models to engage in self-reflection based on dynamic criteria, enabling controlled iterative improvement; on the other hand, we adopt the training paradigm that combines supervised fine-tuning with reinforcement learning to convert evaluation criteria into reward signals, achieving end-to-end optimization. Experimental results demonstrate that the automatically constructed criteria achieve performance gains comparable to human annotations. Writer-R1-4B models trained with this approach outperform baselines across multiple creative writing tasks and surpass some 100B+ parameter open-source models. --- ### Muon Converges under Heavy-Tailed Noise: Nonconvex Hölder-Smooth Empirical Risk Minimization Authors: Hideaki Iiduka Published: 2026-03-16 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2603.15059v2 Muon is a recently proposed optimizer that enforces orthogonality in parameter updates by projecting gradients onto the Stiefel manifold, leading to stable and efficient training in large-scale deep neural networks. Meanwhile, the previously reported results indicated that stochastic noise in practical machine learning may exhibit heavy-tailed behavior, violating the bounded-variance assumption. In this paper, we consider the problem of minimizing a nonconvex Hölder-smooth empirical risk that works well with the heavy-tailed stochastic noise. We then show that Muon converges to a stationary point of the empirical risk under the boundedness condition accounting for heavy-tailed stochastic noise. In addition, we show that Muon converges faster than mini-batch SGD. --- ### Analyzing Error Sources in Global Feature Effect Estimation Authors: Timo Heiß, Coco Bögel, Bernd Bischl, Giuseppe Casalicchio Published: 2026-03-16 Categories: stat.ML, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.15057v2 Global feature effects such as partial dependence (PD) and accumulated local effects (ALE) plots are widely used to interpret black-box models. However, they are only estimates of true underlying effects, and their reliability depends on multiple sources of error. Despite the popularity of global feature effects, these error sources are largely unexplored. In particular, the practically relevant question of whether to use training or holdout data to estimate feature effects remains unanswered. We address this gap by providing a systematic, estimator-level analysis that disentangles sources of bias and variance for PD and ALE. To this end, we derive a mean-squared-error decomposition that separates model bias, estimation bias, model variance, and estimation variance, and analyze their dependence on model characteristics, data selection, and sample size. We validate our theoretical findings through an extensive simulation study across multiple data-generating processes, learners, estimation strategies (training data, validation data, and cross-validation), and sample sizes. Our results reveal that, while using holdout data is theoretically the cleanest, potential biases arising from the training data are empirically negligible and dominated by the impact of the usually higher sample size. The estimation variance depends on both the presence of interactions and the sample size, with ALE being particularly sensitive to the latter. Cross-validation-based estimation is a promising approach that reduces the model variance component, particularly for overfitting models. Our analysis provides a principled explanation of the sources of error in feature effect estimates and offers concrete guidance on choosing estimation strategies when interpreting machine learning models. --- ### Spatio-temporal probabilistic forecast using MMAF-guided learning Authors: Leonardo Bardi, Imma Valentina Curato, Lorenzo Proietti Published: 2026-03-16 Categories: stat.ML, cs.LG, math.ST arXiv: https://arxiv.org/abs/2603.15055v3 We present a theory-guided generalized Bayesian methodology for spatio-temporal raster data, which we use to train an ensemble of stochastic feed-forward neural networks with Gaussian-distributed weights. The methodology incorporates the dependence and causal structure of a spatio-temporal Ornstein-Uhlenbeck process into training and inference by enforcing constraints on the design of the data embedding and the related optimization routine. In inference mode, the networks are employed to generate causal ensemble forecasts by applying different initial conditions at different horizons. We call this workflow MMAF-guided learning. Experiments conducted on both synthetic and real data demonstrate that our forecasts remain calibrated across multiple time horizons. Moreover, we show that on such data, shallow feed-forward architectures can achieve performance comparable to, and in some cases better than, convolutional or diffusion deep learning architectures used in probabilistic forecasting tasks. --- ### Interference-Aware K-Step Reachable Communication in Multi-Agent Reinforcement Learning Authors: Ziyu Cheng, Jinsheng Ren, Zhouxian Jiang, Chenzhihang Li, Rongye Shi, Bin Liang, Jun Yang Published: 2026-03-16 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.15054v1 Effective communication is pivotal for addressing complex collaborative tasks in multi-agent reinforcement learning (MARL). Yet, limited communication bandwidth and dynamic, intricate environmental topologies present significant challenges in identifying high-value communication partners. Agents must consequently select collaborators under uncertainty, lacking a priori knowledge of which partners can deliver task-critical information. To this end, we propose Interference-Aware K-Step Reachable Communication (IA-KRC), a novel framework that enhances cooperation via two core components: (1) a K-Step reachability protocol that confines message passing to physically accessible neighbors, and (2) an interference-prediction module that optimizes partner choice by minimizing interference while maximizing utility. Compared to existing methods, IA-KRC enables substantially more persistent and efficient cooperation despite environmental interference. Comprehensive evaluations confirm that IA-KRC achieves superior performance compared to state-of-the-art baselines, while demonstrating enhanced robustness and scalability in complex topological and highly dynamic multi-agent scenarios. --- ### Thinking in Latents: Adaptive Anchor Refinement for Implicit Reasoning in LLMs Authors: Disha Sheshanarayana, Rajat Subhra Pal, Manjira Sinha, Tirthankar Dasgupta Published: 2026-03-16 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.15051v1 Token-level Chain-of-Thought (CoT) prompting has become a standard way to elicit multi-step reasoning in large language models (LLMs), especially for mathematical word problems. However, generating long intermediate traces increases output length and inference cost, and can be inefficient when the model could arrive at the correct answer without extensive verbalization. This has motivated latent-space reasoning approaches that shift computation into hidden representations and only emit a final answer. Yet, many latent reasoning methods depend on a fixed number of latent refinement steps at inference, adding another hyperparameter that must be tuned across models and datasets to balance accuracy and efficiency. We introduce AdaAnchor, a latent reasoning framework that performs silent iterative computation by refining a set of latent anchor vectors attached to the input. AdaAnchor further incorporates an adaptive halting mechanism that monitors anchor stability across iterations and terminates refinement once the anchor dynamics converge, allocating fewer steps to easier instances while reserving additional refinement steps for harder ones under a shared maximum-step budget. Our empirical evaluation across three mathematical word-problem benchmarks shows that AdaAnchor with adaptive halting yields accuracy gains of up to 5% over fixed-step latent refinement while reducing average latent refinement steps by 48-60% under the same maximum-step budget. Compared to standard reasoning baselines, AdaAnchor achieves large reductions in generated tokens (92-93%) by moving computation into silent latent refinement, offering a different accuracy-efficiency trade-off with substantially lower output-token usage. --- ### SRL-MAD: Structured Residual Latents for One-Class Morphing Attack Detection Authors: Diogo J. Paulo, Hugo Proença, João C. Neves Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15050v1 Face morphing attacks represent a significant threat to biometric systems as they allow multiple identities to be combined into a single face. While supervised morphing attack detection (MAD) methods have shown promising performance, their reliance on attack-labeled data limits generalization to unseen morphing attacks. This has motivated increasing interest in one-class MAD, where models are trained exclusively on bona fide samples and are expected to detect unseen attacks as deviations from the normal facial structure. In this context, we introduce SRL-MAD, a one-class single-image MAD that uses structured residual Fourier representations for open-set morphing attack detection. Starting from a residual frequency map that suppresses image-specific spectral trends, we preserve the two-dimensional organization of the Fourier domain through a ring-based representation and replace azimuthal averaging with a learnable ring-wise spectral projection. To further encode domain knowledge about where morphing artifacts arise, we impose a frequency-informed inductive bias by organizing spectral evidence into low, mid, and high-frequency bands and learning cross-band interactions. These structured spectral features are mapped into a latent space designed for direct scoring, avoiding the reliance on reconstruction errors. Extensive evaluation on FERET-Morph, FRLL-Morph, and MorDIFF demonstrates that SRL-MAD consistently outperforms recent one-class and supervised MAD models. Overall, our results show that learning frequency-aware projections provides a more discriminative alternative to azimuthal spectral summarization for one-class morphing attack detection. --- ### CrossADR: enhancing adverse drug reactions prediction for combination pharmacotherapy with cross-layer feature integration and cross-level associative learning Authors: Y. Cheung Published: 2026-03-16 Categories: cs.LG, math.AG arXiv: https://arxiv.org/abs/2603.15047v1 Combination pharmacotherapy offers substantial therapeutic advantages but also poses substantial risks of adverse drug reactions (ADRs). The accurate prediction of ADRs with interpretable computational methods is crucial for clinical safety management, drug development, and precision medicine. However, managing ADRs remains a challenge due to the vast search space of drug combinations and the complexity of physiological responses. Current graph-based architectures often struggle to effectively integrate multi-scale biological information and frequently rely on fixed association matrices, which limits their ability to capture dynamic organ-level dependencies and generalize across diverse datasets. Here we propose CrossADR, a hierarchical framework for organ-level ADR prediction through cross-layer feature integration and cross-level associative learning. It incorporates a gated-residual-flow graph neural network to fuse multi-scale molecular features and utilizes a learnable ADR embedding space to dynamically capture latent biological correlations across 15 organ systems. Systematic evaluation on the newly constructed CrossADR-Dataset-covering 1,376 drugs and 946,000 unique combinations-demonstrates that CrossADR consistently achieves state-of-the-art performance across 80 distinct experimental scenarios and provides high-resolution insights into drug-related protein protein interactions and pathways. Overall, CrossADR represents a robust tool for cross-scale biomedical information integration, cross-layer feature integration as well as cross-level associative learning, and can be effectively utilized to prevent ADRs in clinical decision-making. --- ### AnoleVLA: Lightweight Vision-Language-Action Model with Deep State Space Models for Mobile Manipulation Authors: Yusuke Takagi, Motonari Kambara, Daichi Yashima, Koki Seno, Kento Tokura, Komei Sugiura Published: 2026-03-16 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.15046v1 In this study, we address the problem of language-guided robotic manipulation, where a robot is required to manipulate a wide range of objects based on visual observations and natural language instructions. This task is essential for service robots that operate in human environments, and requires safety, efficiency, and task-level generality. Although Vision-Language-Action models (VLAs) have demonstrated strong performance for this task, their deployment in resource-constrained environments remains challenging because of the computational cost of standard transformer backbones. To overcome this limitation, we propose AnoleVLA, a lightweight VLA that uses a deep state space model to process multimodal sequences efficiently. The model leverages its lightweight and fast sequential state modeling to process visual and textual inputs, which allows the robot to generate trajectories efficiently. We evaluated the proposed method in both simulation and physical experiments. Notably, in real-world evaluations, AnoleVLA outperformed a representative large-scale VLA by 21 points for the task success rate while achieving an inference speed approximately three times faster. --- ### Prompt Readiness Levels (PRL): a maturity scale and scoring framework for production grade prompt assets Authors: Sebastien Guinard Published: 2026-03-16 Categories: cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2603.15044v1 Prompt engineering has become a production critical component of generative AI systems. However, organizations still lack a shared, auditable method to qualify prompt assets against operational objectives, safety constraints, and compliance requirements. This paper introduces Prompt Readiness Levels (PRL), a nine level maturity scale inspired by TRL, and the Prompt Readiness Score (PRS), a multidimensional scoring method with gating thresholds designed to prevent weak link failure modes. PRL/PRS provide an original, structured and methodological framework for governing prompt assets specification, testing, traceability, security evaluation, and deployment readiness enabling valuation of prompt engineering through reproducible qualification decisions across teams and industries. --- ### GUI-CEval: A Hierarchical and Comprehensive Chinese Benchmark for Mobile GUI Agents Authors: Yang Li, Yuchen Liu, Haoyu Lu, Zhiqiang Xia, Hongzhen Wang, Kaiyang Han, Changpeng Yang, Jinyang Wu, Jiaming Xu, Runyu Shi, Ying Huang Published: 2026-03-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.15039v1 Recent progress in Multimodal Large Language Models (MLLMs) has enabled mobile GUI agents capable of visual perception, cross-modal reasoning, and interactive control. However, existing benchmarks are largely English-centric and fail to capture the linguistic and interaction characteristics of the Chinese mobile ecosystem. They also focus on isolated skills such as GUI grounding or offline agent, lacking a unified and fine-grained framework to assess the full capability chain from perception to execution. To address this gap, we introduce GUI-CEval, the first comprehensive benchmark for Chinese mobile GUI agents, built entirely on physical device environments. GUI-CEval spans 201 mainstream apps across four device types and adopts a two-level structure that evaluates both atomic abilities and realistic application-level performance along five dimensions: perception, planning, reflection, execution, and evaluation. All data are collected and verified through multi-stage manual processes to ensure authenticity and reproducibility. Extensive experiments on 20 representative MLLMs and multi-agent systems show that while models such as Qwen2.5-VL and UI-TARS perform competitively, most MLLMs still exhibit clear weaknesses in reflective decision-making and post-action self-evaluation, limiting their reliability in real-world interactions. We hope GUI-CEval provides a comprehensive and interpretable benchmark to guide capability diagnosis and advance the development of Chinese mobile GUI agents. --- ### A convolutional autoencoder and neural ODE surrogate modeling framework applied to transient counterflow flames Authors: Mert Yakup Baykan, Weitao Liu, Mohammad Rafi Malik, Thorsten Zirwes, Andreas Kronenburg, Hong G. Im, Dong-hyuk Shin Published: 2026-03-16 Categories: physics.flu-dyn, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2603.15038v2 A novel convolutional autoencoder and neural ODE (CAE-NODE) framework is proposed for a reduced-order model (ROM) applied to transient 2D counterflow flames, as an extension of AE-NODE methods in homogeneous reactive systems to spatially resolved flows. The multidimensional thermochemical fields (256 x 256 grid, 21 variables) obtained from direct numerical simulations (DNS) are used in training the CAE, where convolutional layers learned the underlying spatial correlations, allowing the CAE to construct an unsupervised 3D latent manifold that is physically meaningful, smooth, and continuous in time. This results in a compression ratio of over 400,000 times. The NODE then subsequently learns the continuous-time dynamics on the latent manifold, enabling the prediction of the full temporal evolution of the flames by integrating forward in time from an initial condition. The results demonstrate that the CAE-NODE can accurately capture the entire transient process, including ignition, flame propagation, and the gradual transition to a non-premixed condition, with excellent agreement with the DNS, while adhering to conservation principles at virtually no computational cost compared to the reference DNS. Predictions remain accurate at strain rates outside the training range. Moreover, despite being unsupervised, the learned latent manifold is highly correlated with the flame-state descriptors such as the progress variable, mixture fraction, and the scalar dissipation rate. This study, for the first time, highlights the potential of CAE-NODE for surrogate modeling of unsteady dynamics of multi-dimensional reacting flows. --- ### A convolutional autoencoder and neural ODE framework for surrogate modeling of transient counterflow flames Authors: Mert Yakup Baykan, Weitao Liu, Thorsten Zirwes, Andreas Kronenburg, Hong G. Im, Dong-hyuk Shin Published: 2026-03-16 Categories: physics.flu-dyn, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2603.15038v1 A novel convolutional autoencoder neural ODE (CAE-NODE) framework is proposed for a reduced-order model (ROM) of transient 2D counterflow flames, as an extension of AE-NODE methods in homogeneous reactive systems to spatially resolved flows. The spatial correlations of the multidimensional fields are extracted by the convolutional layers, allowing CAE to autonomously construct a physically consistent 6D continuous latent manifold by compressing high-fidelity 2D snapshots (256x256 grid, 21 variables) by over 100,000 times. The NODE is subsequently trained to describe the continuous-time dynamics on the non-linear manifold, enabling the prediction of the full temporal evolution of the flames by integrating forward in time from an initial condition. The results demonstrate that the network can accurately capture the entire transient process, including ignition, flame propagation, and the gradual transition to a non-premixed condition, with relative errors less than ~2% for major species. This study, for the first time, highlights the potential of CAE-NODE for surrogate modeling of unsteady dynamics of multi-dimensional reacting flows. --- ### Interpretable Predictability-Based AI Text Detection: A Replication Study Authors: Adam Skurla, Dominik Macko, Jakub Simko Published: 2026-03-16 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.15034v2 This paper replicates and extends the system used in the AuTexTification shared task for authorship attribution of machine-generated texts. Exact replication was not possible because of differences in data splits, model availability, and implementation details, which we document as a case study in reproducibility. We tested newer multilingual language models (mDeBERTa-v3-base, Qwen, mGPT) and added 26 document-level stylometric features, using ablation, permutation importance, and SHAP analysis to assess feature influence. A single shared configuration was applied to both English and Spanish across Subtask 1 and Subtask 2. Averaged over three random seeds, the shared multilingual configuration performs comparably to or better than the language-specific baseline, with the clearest gains on model attribution (Subtask 2). The additional stylometric features yield small improvements, led by lexical diversity, but their contribution falls within seed variance once predictability-based probabilities are included, which remain the dominant signal. The study also shows that clear documentation is important for reliable replication and fair comparison of systems. --- ### Interpretable Predictability-Based AI Text Detection: A Replication Study Authors: Adam Skurla, Dominik Macko, Jakub Simko Published: 2026-03-16 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.15034v1 This paper replicates and extends the system used in the AuTexTification 2023 shared task for authorship attribution of machine-generated texts. First, we tried to reproduce the original results. Exact replication was not possible because of differences in data splits, model availability, and implementation details. Next, we tested newer multilingual language models and added 26 document-level stylometric features. We also applied SHAP analysis to examine which features influence the model's decisions. We replaced the original GPT-2 models with newer generative models such as Qwen and mGPT for computing probabilistic features. For contextual representations, we used mDeBERTa-v3-base and applied the same configuration to both English and Spanish. This allowed us to use one shared configuration for Subtask 1 and Subtask 2. Our experiments show that the additional stylometric features improve performance in both tasks and both languages. The multilingual configuration achieves the results that are comparable to or better than language-specific models. The study also shows that clear documentation is important for reliable replication and fair comparison of systems. --- ### Models Designed to Forget: Machine Unlearning via Key Deletion Authors: Sonia Laguna, Jorge da Silva Goncalves, Moritz Vandenhirtz, Alain Ryser, Irene Cannistraci, Julia E. Vogt Published: 2026-03-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.15033v4 Machine unlearning for vision models is rapidly becoming a practical requirement, driven by privacy regulations, data errors, and the need to remove harmful or corrupted training images. Despite this, most existing approximate unlearning methods tackle the problem from a post-hoc perspective. They attempt to erase the influence of targeted samples through parameter updates that typically require access to the full training data. This creates a mismatch with real deployment scenarios where unlearning requests can be anticipated, revealing a fundamental limitation of post-hoc approaches. We motivate unlearning by design, a novel paradigm for approximate methods in which models are directly trained to support forgetting as an inherent architectural capability. We instantiate this idea with Machine UNlearning via KEY deletion (MUNKEY), a memory-augmented transformer that decouples instance-specific memorization from model weights. Here, unlearning corresponds to removing the instance-identifying key, enabling zero-shot forgetting without weight updates or access to the original samples or labels. Across natural image benchmarks, fine-grained visual recognition, and medical datasets, MUNKEY outperforms all post-hoc baselines. Our results establish that unlearning by design enables fast, deployment-oriented unlearning while preserving predictive performance. --- ### Rethinking Machine Unlearning: Models Designed to Forget via Key Deletion Authors: Sonia Laguna, Jorge da Silva Goncalves, Moritz Vandenhirtz, Alain Ryser, Irene Cannistraci, Julia E. Vogt Published: 2026-03-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.15033v3 Machine unlearning is rapidly becoming a practical requirement, driven by privacy regulations, data errors, and the need to remove harmful or corrupted training samples. Despite this, most existing methods tackle the problem purely from a post-hoc perspective. They attempt to erase the influence of targeted training samples through parameter updates that typically require access to the full training data. This creates a mismatch with real deployment scenarios where unlearning requests can be anticipated, revealing a fundamental limitation of post-hoc approaches. We propose unlearning by design, a novel paradigm in which models are directly trained to support forgetting as an inherent capability. We instantiate this idea with Machine UNlearning via KEY deletion (MUNKEY), a memory augmented transformer that decouples instance-specific memorization from model weights. Here, unlearning corresponds to removing the instance-identifying key, enabling direct zero-shot forgetting without weight updates or access to the original samples or labels. Across natural image benchmarks, fine-grained recognition, and medical datasets, MUNKEY outperforms all post-hoc baselines. Our results establish that unlearning by design enables fast, deployment-oriented unlearning while preserving predictive performance. --- ### Attention Residuals Authors: Kimi Team, Guangyu Chen, Yu Zhang, Jianlin Su, Weixin Xu, Siyuan Pan, Yaoyu Wang, Yucheng Wang, Guanduo Chen, Bohong Yin, Yutian Chen, Junjie Yan et al. Published: 2026-03-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.15031v1 Residual connections with PreNorm are standard in modern LLMs, yet they accumulate all layer outputs with fixed unit weights. This uniform aggregation causes uncontrolled hidden-state growth with depth, progressively diluting each layer's contribution. We propose Attention Residuals (AttnRes), which replaces this fixed accumulation with softmax attention over preceding layer outputs, allowing each layer to selectively aggregate earlier representations with learned, input-dependent weights. To address the memory and communication overhead of attending over all preceding layer outputs for large-scale model training, we introduce Block AttnRes, which partitions layers into blocks and attends over block-level representations, reducing the memory footprint while preserving most of the gains of full AttnRes. Combined with cache-based pipeline communication and a two-phase computation strategy, Block AttnRes becomes a practical drop-in replacement for standard residual connections with minimal overhead. Scaling law experiments confirm that the improvement is consistent across model sizes, and ablations validate the benefit of content-dependent depth-wise selection. We further integrate AttnRes into the Kimi Linear architecture (48B total / 3B activated parameters) and pre-train on 1.4T tokens, where AttnRes mitigates PreNorm dilution, yielding more uniform output magnitudes and gradient distribution across depth, and improves downstream performance across all evaluated tasks. --- ### VTC-Bench: Evaluating Agentic Multimodal Models via Compositional Visual Tool Chaining Authors: Xuanyu Zhu, Yuhao Dong, Rundong Wang, Yang Shi, Zhipeng Wu, Yinlun Peng, YiFan Zhang, Yihang Lou, Yuanxing Zhang, Ziwei Liu, Yan Bai, Yuan Zhou Published: 2026-03-16 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.15030v2 Recent advancements extend Multimodal Large Language Models (MLLMs) beyond standard visual question answering to utilizing external tools for advanced visual tasks. Despite this progress, precisely executing and effectively composing diverse tools for complex tasks remain persistent bottleneck. Constrained by sparse tool-sets and simple tool-use trajectories, existing benchmarks fail to capture complex and diverse tool interactions, falling short in evaluating model performance under practical, real-world conditions. To bridge this gap, we introduce VisualToolChain-Bench(VTC-Bench), a comprehensive benchmark designed to evaluate tool-use proficiency in MLLMs. To align with realistic computer vision pipelines, our framework features 32 diverse OpenCV-based visual operations. This rich tool-set enables extensive combinations, allowing VTC-Bench to rigorously assess multi-tool composition and long-horizon, multi-step plan execution. For precise evaluation, we provide 680 curated problems structured across a nine-category cognitive hierarchy, each with ground-truth execution trajectories. Extensive experiments on 19 leading MLLMs reveal critical limitations in current models' visual agentic capabilities. Specifically, models struggle to adapt to diverse tool-sets and generalize to unseen operations, with the leading model Gemini-3.0-Pro only achieving 51% on our benchmark. Furthermore, multi-tool composition remains a persistent challenge. When facing complex tasks, models struggle to formulate efficient execution plans, relying heavily on a narrow, suboptimal subset of familiar functions rather than selecting the optimal tools. By identifying these fundamental challenges, VTC-Bench establishes a rigorous baseline to guide the development of more generalized visual agentic models. ---