# Research Papers — 2026-03 · bundle page 197 of 243 > Full-text markdown of papers #9801–9850 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/197 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/196.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/198.md) ### GreenRFM: Learning a resource-efficient radiology vision-language foundation model via supervision-centric pre-training Authors: Yingtai Li, Shuai Ming, Qiuli Wang, Mingyue Zhao, Hongchun Zhang, Yuhe Tian, Haoran Lai, Rongsheng Wang, Rui Zhou, Rundong Wang, Yujia Li, Zhiyang He et al. Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06467v3 Radiology foundation models (RFMs) have largely inherited the scale-first recipe of natural-image vision--language pre-training. This recipe is difficult to deploy in 3D radiology, where training corpora are smaller, reports vary across institutions, and receiving hospitals often need local adaptation under privacy and compute constraints. We ask whether routine radiology reports can instead be converted into auditable diagnostic supervision that shapes the image encoder, text encoder, aligned space, and local-adaptation procedure. We develop GreenRFM, a supervision-centric pre-training framework organized around four empirical principles: More distilled, Ubiquitous, Semantic-enforcing, and Task-aligning (MUST) supervision. These principles convert noisy reports into structured diagnostic signals and use them to learn discriminative unimodal encoders plus an aligned image--text space for diagnosis-centered multimodal use. GreenRFM requires 24 GPU-hours on a single 24GB GPU (lightweight variant: 6GB VRAM, 4~hours) and reaches a zero-shot CT-RATE AUC of 84.8. Evaluations using more than 200,000 volumes from six institutions and two modalities show transfer to private clinical cohorts and to musculoskeletal MRI. On a local institutional cohort, computationally feasible retraining raises macro-AUC from 70.5 to 82.1. The aligned space also improves hepatocellular-carcinoma microvascular-invasion prediction and trans-arterial chemoembolization response analysis over established clinical scores. These results support supervision-centric pre-training as a practical route to resource-efficient, locally adaptable, diagnosis-centered radiology vision--language representations. --- ### Bayesian Additive Distribution Regression Authors: Antonio R. Linero, Soumyabrata Bose, Jared Murray Published: 2026-03-06 Categories: stat.ME, stat.ML arXiv: https://arxiv.org/abs/2603.06462v1 Distribution regression, where the goal is to predict a scalar response from a distribution-valued predictor, arises naturally in settings where observations are grouped and outcomes depend on group-level characteristics rather than on individual measurements. We introduce DistBART, a Bayesian nonparametric approach to distribution regression that models the regression function as a linear functional with the Riesz representer assigned a Bayesian additive regression trees (BART) prior. We argue that shallow decision tree ensembles encode reasonable inductive biases for tabular data, making them appropriate in settings where the functional depends primarily on low-dimensional marginals of the distributions. We show this both empirically on synthetic and real data and theoretically through an adaptive posterior concentration result. We also establish connections to kernel methods, and use this connection to motivate variants of DistBART that can learn nonlinear functionals. To enable scalability to large datasets, we develop a random-feature approximation that samples trees from the BART prior and reduces inference to sparse Bayesian linear regression, achieving computational efficiency while retaining uncertainty quantification. --- ### Do Foundation Models Know Geometry? Probing Frozen Features for Continuous Physical Measurement Authors: Yakov Pyotr Shkolnikov Published: 2026-03-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.06459v1 Vision-language models encode continuous geometry that their text pathway fails to express: a 6,000-parameter linear probe extracts hand joint angles at 6.1 degrees MAE from frozen features, while the best text output achieves only 20.0 degrees -- a 3.3x bottleneck. LoRA fine-tuning (r=16, 2,000 images) narrows this gap to 6.5 degrees, providing evidence for a pathway-training deficit rather than a representational one. Training objective determines accuracy more than architecture: five encoders spanning self-supervised, contrastive, and hybrid paradigms converge to statistically equivalent accuracy (R^2 approximately 0.55, TOST-equivalent at delta=0.03) despite sharing as little as CKA=0.41 representational similarity -- functional convergence without representational convergence. Autoregressive generation damages geometric fidelity, but the damage originates in the generation process, not in language alignment: Qwen2.5-VL's LLM layers actually improve probe accuracy over its raw vision encoder. Layer-wise analysis reveals a universal mid-network accuracy peak across all architectures, with attention heads in layers 18-22 carrying disproportionate geometric signal. These findings enable a single frozen backbone to function as a multi-task geometric sensor through lightweight probes, without fine-tuning or text generation. --- ### Training Flow Matching: The Role of Weighting and Parameterization Authors: Anne Gagneux, Ségolène Martin, Rémi Gribonval, Mathurin Massias Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06454v3 We study the training objectives of denoising-based generative models, with a particular focus on loss weighting and output parameterization, including noise-, clean image-, and velocity-based formulations. Through a systematic numerical study, we analyze how these training choices interact with the intrinsic dimensionality of the data manifold, model architecture, and dataset size. Our experiments span synthetic datasets with controlled geometry as well as image data, and compare training objectives using quantitative metrics for denoising accuracy (PSNR across noise levels) and generative quality (FID). Rather than proposing a new method, our goal is to disentangle the various factors that matter when training a flow matching model, in order to provide practical insights on design choices. --- ### Training Flow Matching: The Role of Weighting and Parameterization Authors: Anne Gagneux, Ségolène Martin, Rémi Gribonval, Mathurin Massias Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06454v2 We study the training objectives of denoising-based generative models, with a particular focus on loss weighting and output parameterization, including noise-, clean image-, and velocity-based formulations. Through a systematic numerical study, we analyze how these training choices interact with the intrinsic dimensionality of the data manifold, model architecture, and dataset size. Our experiments span synthetic datasets with controlled geometry as well as image data, and compare training objectives using quantitative metrics for denoising accuracy (PSNR across noise levels) and generative quality (FID). Rather than proposing a new method, our goal is to disentangle the various factors that matter when training a flow matching model, in order to provide practical insights on design choices. --- ### Pinterest Canvas: Large-Scale Image Generation at Pinterest Authors: Yu Wang, Eric Tzeng, Raymond Shiau, Jie Yang, Dmitry Kislyuk, Charles Rosenberg Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06453v2 While recent image generation models demonstrate a remarkable ability to handle a wide variety of image generation tasks, this flexibility makes them hard to control via prompting or simple inference adaptation alone, rendering them unsuitable for use cases with strict product requirements. In this paper, we introduce Pinterest Canvas, our large-scale image generation system built to support image editing and enhancement use cases at Pinterest. Canvas is first trained on a diverse, multimodal dataset to produce a foundational diffusion model with broad image-editing capabilities. However, rather than relying on one generic model to handle every downstream task, we instead rapidly fine-tune variants of this base model on task-specific datasets, producing specialized models for individual use cases. We describe key components of Canvas and summarize our best practices for dataset curation, training, and inference. We also showcase task-specific variants through case studies on background enhancement and aspect-ratio outpainting, highlighting how we tackle their specific product requirements. Online A/B experiments demonstrate that our enhanced images receive a significant 18.0% and 12.5% engagement lift, respectively, and comparisons with human raters further validate that our models outperform third-party models on these tasks. Finally, we showcase other Canvas variants, including multi-image scene synthesis and image-to-video generation, demonstrating that our approach can generalize to a wide variety of potential downstream tasks. --- ### CaTok: Taming Mean Flows for One-Dimensional Causal Image Tokenization Authors: Yitong Chen, Zuxuan Wu, Xipeng Qiu, Yu-Gang Jiang Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06449v1 Autoregressive (AR) language models rely on causal tokenization, but extending this paradigm to vision remains non-trivial. Current visual tokenizers either flatten 2D patches into non-causal sequences or enforce heuristic orderings that misalign with the "next-token prediction" pattern. Recent diffusion autoencoders similarly fall short: conditioning the decoder on all tokens lacks causality, while applying nested dropout mechanism introduces imbalance. To address these challenges, we present CaTok, a 1D causal image tokenizer with a MeanFlow decoder. By selecting tokens over time intervals and binding them to the MeanFlow objective, as illustrated in Fig. 1, CaTok learns causal 1D representations that support both fast one-step generation and high-fidelity multi-step sampling, while naturally capturing diverse visual concepts across token intervals. To further stabilize and accelerate training, we propose a straightforward regularization REPA-A, which aligns encoder features with Vision Foundation Models (VFMs). Experiments demonstrate that CaTok achieves state-of-the-art results on ImageNet reconstruction, reaching 0.75 FID, 22.53 PSNR and 0.674 SSIM with fewer training epochs, and the AR model attains performance comparable to leading approaches. --- ### What if? Emulative Simulation with World Models for Situated Reasoning Authors: Ruiping Liu, Yufan Chen, Yuheng Zhang, Junwei Zheng, Kunyu Peng, Chengzhi Wu, Chenguang Huang, Di Wen, Jiaming Zhang, Kailun Yang, Rainer Stiefelhagen Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06445v3 Situated reasoning often relies on active exploration, yet in many real-world scenarios such exploration is infeasible due to physical constraints of robots or safety concerns of visually impaired users. Given only a limited observation, can an agent mentally simulate a future trajectory toward a target situation and answer spatial what-if questions? We introduce WanderDream, the first large-scale dataset designed for the emulative simulation of mental exploration, enabling models to reason without active exploration. WanderDream-Gen comprises 15.8K panoramic videos across 1,088 real scenes from HM3D, ScanNet++, and real-world captures, depicting imagined trajectories from current viewpoints to target situations. WanderDream-QA contains 158K question-answer pairs, covering starting states, paths, and end states along each trajectory to comprehensively evaluate exploration-based reasoning. Extensive experiments with world models and MLLMs demonstrate (1) that mental exploration is essential for situated reasoning, (2) that world models achieve compelling performance on WanderDream-Gen, (3) that imagination substantially facilitates reasoning on WanderDream-QA, and (4) that WanderDream data exhibit remarkable transferability to real-world scenarios. --- ### Prosodic Boundary-Aware Streaming Generation for LLM-Based TTS with Streaming Text Input Authors: Changsong Liu, Tianrui Wang, Ye Ni, Yizhou Peng, Eng Siong Chng Published: 2026-03-06 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2603.06444v1 Streaming TTS that receives streaming text is essential for interactive systems, yet this scheme faces two major challenges: unnatural prosody due to missing lookahead and long-form collapse due to unbounded context. We propose a prosodic-boundary-aware post-training strategy, adapting a pretrained LLM-based TTS model using weakly time-aligned data. Specifically, the model is adapted to learn early stopping at specified content boundaries when provided with limited future text. During inference, a sliding-window prompt carries forward previous text and speech tokens, ensuring bounded context and seamless concatenation. Evaluations show our method outperforms CosyVoice-Style interleaved baseline in both short and long-form scenarios. In long-text synthesis, especially, it achieves a 66.2% absolute reduction in word error rate (from 71.0% to 4.8%) and increases speaker and emotion similarity by 16.1% and 1.5% relatively, offering a robust solution for streaming TTS with incremental text. --- ### Toward Generative Quantum Utility via Correlation-Complexity Map Authors: Chen-Yu Liu, Leonardo Placidi, Eric Brunner, Enrico Rinaldi Published: 2026-03-06 Categories: cs.LG, quant-ph arXiv: https://arxiv.org/abs/2603.06440v2 We study a practical question in generative quantum machine learning: given a classical dataset, can we determine, before training, whether it is well suited to a quantum generative model? We focus on a class of quantum circuits known as instantaneous quantum polynomial-time (IQP) circuits, whose output distributions are widely believed to be difficult to sample from using classical methods. These circuits are used to build our quantum generative models. We introduce a Correlation-Complexity Map, a simple diagnostic built from two quantities computed from data samples. The first measures how closely the dataset's spectral correlation patterns resemble those naturally produced by IQP circuits, while the second quantifies how much of the dataset's structural correlation cannot be captured by simple pairwise models. In other words, we can estimate beforehand how well a dataset can be approximated by our model family and also how complex its correlations are, indicating possible failures of classical models. Applying this framework, we identify turbulence data as a promising target for quantum generative modeling. Guided by this analysis, we use a latent-parameter adaptation scheme that reuses a compact IQP circuit over a temporal sequence by learning and interpolating a low-dimensional latent trajectory, and observe competitive performance against classical baselines in a low-data, low-parameter regime. These results suggest that dataset-level diagnostics can help prioritize problems where quantum generative models are most likely to be useful, with improvements in data and parameter efficiency. --- ### Certified and accurate computation of function space norms of deep neural networks Authors: Johannes Gründler, Moritz Maibaum, Philipp Petersen Published: 2026-03-06 Categories: math.NA, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.06431v2 Neural network methods for PDEs require reliable error control in function space norms. However, trained neural networks can typically only be probed at a finite number of point values. Without strong assumptions, point evaluations alone do not provide enough information to derive tight deterministic and guaranteed bounds on function space norms. In this work, we move beyond a purely black-box setting and exploit the neural network structure directly. We present a framework for the certified and accurate computation of integral quantities of neural networks, including Lebesgue and Sobolev norms, by combining interval arithmetic enclosures on axis-aligned boxes with adaptive marking/refinement and quadrature-based aggregation. On each box, we compute guaranteed lower and upper bounds for function values and derivatives, and propagate these local certificates to global lower and upper bounds for the target integrals. Our analysis provides a general convergence theorem for such certified adaptive quadrature procedures and instantiates it for function values, Jacobians, and Hessians, yielding certified computation of $L^p$, $W^{1,p}$, and $W^{2,p}$ norms. We further show how these ingredients lead to practical certified bounds for PINN interior residuals. Numerical experiments illustrate the accuracy and practical behavior of the proposed methods. --- ### Abductive Reasoning with Syllogistic Forms in Large Language Models Authors: Hirohiko Abe, Risako Ando, Takanobu Morishita Kentaro Ozeki, Koji Mineshima, Mitsuhiro Okada Published: 2026-03-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.06428v1 Research in AI using Large-Language Models (LLMs) is rapidly evolving, and the comparison of their performance with human reasoning has become a key concern. Prior studies have indicated that LLMs and humans share similar biases, such as dismissing logically valid inferences that contradict common beliefs. However, criticizing LLMs for these biases might be unfair, considering our reasoning not only involves formal deduction but also abduction, which draws tentative conclusions from limited information. Abduction can be regarded as the inverse form of syllogism in its basic structure, that is, a process of drawing a minor premise from a major premise and conclusion. This paper explores the accuracy of LLMs in abductive reasoning by converting a syllogistic dataset into one suitable for abduction. It aims to investigate whether the state-of-the-art LLMs exhibit biases in abduction and to identify potential areas for improvement, emphasizing the importance of contextualized reasoning beyond formal deduction. This investigation is vital for advancing the understanding and application of LLMs in complex reasoning tasks, offering insights into bridging the gap between machine and human cognition. --- ### CLoPA: Continual Low Parameter Adaptation of Interactive Segmentation for Medical Image Annotation Authors: Parhom Esmaeili, Chayanin Tangwiriyasakul, Eli Gibson, Sebastien Ourselin, M. Jorge Cardoso Published: 2026-03-06 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.06426v1 Interactive segmentation enables clinicians to guide annotation, but existing zero-shot models like nnInteractive fail to consistently reach expert-level performance across diverse medical imaging tasks. Because annotation campaigns produce a growing stream of task-specific labelled data, online adaptation of the segmentation model is a natural complement to zero-shot inference. We propose CLoPA, a continual adaptation strategy that tunes a small fraction of nnInteractive's parameters on the annotation cache, triggered by lightweight episode scheduling. CLoPA requires no new parameters or changes to the inference pipeline, and operates entirely within the existing annotation workflow. Across eight Medical Segmentation Decathlon tasks spanning diverse anatomical targets and imaging characteristics, CLoPA rapidly elevates performance to expert-level, even for tasks where nnInteractive previously failed, with the majority of gains realised after a single training episode. We show that the benefits of tuning different parameter groups depends on task characteristics and data regimes. Also, that for targets with complex geometries (e.g., hepatic vessels), instance normalisation and low-level feature tuning saturates, suggesting a need for deeper feature-representation alignment in the most challenging scenarios. --- ### From Prompting to Preference Optimization: A Comparative Study of LLM-based Automated Essay Scoring Authors: Minh Hoang Nguyen, Vu Hoang Pham, Xuan Thanh Huynh, Phuc Hong Mai, Vinh The Nguyen, Quang Nhut Huynh, Huy Tien Nguyen, Tung Le Published: 2026-03-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.06424v1 Large language models (LLMs) have recently reshaped Automated Essay Scoring (AES), yet prior studies typically examine individual techniques in isolation, limiting understanding of their relative merits for English as a Second Language (L2) writing. To bridge this gap, we presents a comprehensive comparison of major LLM-based AES paradigms on IELTS Writing Task~2. On this unified benchmark, we evaluate four approaches: (i) encoder-based classification fine-tuning, (ii) zero- and few-shot prompting, (iii) instruction tuning and Retrieval-Augmented Generation (RAG), and (iv) Supervised Fine-Tuning combined with Direct Preference Optimization (DPO) and RAG. Our results reveal clear accuracy-cost-robustness trade-offs across methods, the best configuration, integrating k-SFT and RAG, achieves the strongest overall results with F1-Score 93%. This study offers the first unified empirical comparison of modern LLM-based AES strategies for English L2, promising potential in auto-grading writing tasks. Code is public at https://github.com/MinhNguyenDS/LLM_AES-EnL2 --- ### Non-invasive Growth Monitoring of Small Freshwater Fish in Home Aquariums via Stereo Vision Authors: Clemens Seibold, Anna Hilsmann, Peter Eisert Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06421v1 Monitoring fish growth behavior provides relevant information about fish health in aquaculture and home aquariums. Yet, monitoring fish sizes poses different challenges, as fish are small and subject to strong refractive distortions in aquarium environments. Image-based measurement offers a practical, non-invasive alternative that allows frequent monitoring without disturbing the fish. In this paper, we propose a non-invasive refraction-aware stereo vision method to estimate fish length in aquariums. Our approach uses a YOLOv11-Pose network to detect fish and predict anatomical keypoints on the fish in each stereo image. A refraction-aware epipolar constraint accounting for the air-glass-water interfaces enables robust matching, and unreliable detections are removed using a learned quality score. A subsequent refraction-aware 3D triangulation recovers 3D keypoints, from which fish length is measured. We validate our approach on a new stereo dataset of endangered Sulawesi ricefish captured under aquarium-like conditions and demonstrate that filtering low-quality detections is essential for accurate length estimation. The proposed system offers a simple and practical solution for non-invasive growth monitoring and can be easily applied in home aquariums. --- ### Evaluation of Deontic Conditional Reasoning in Large Language Models: The Case of Wason's Selection Task Authors: Hirohiko Abe, Kentaro Ozeki, Risako Ando, Takanobu Morishita, Koji Mineshima, Mitsuhiro Okada Published: 2026-03-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.06416v1 As large language models (LLMs) advance in linguistic competence, their reasoning abilities are gaining increasing attention. In humans, reasoning often performs well in domain specific settings, particularly in normative rather than purely formal contexts. Although prior studies have compared LLM and human reasoning, the domain specificity of LLM reasoning remains underexplored. In this study, we introduce a new Wason Selection Task dataset that explicitly encodes deontic modality to systematically distinguish deontic from descriptive conditionals, and use it to examine LLMs' conditional reasoning under deontic rules. We further analyze whether observed error patterns are better explained by confirmation bias (a tendency to seek rule-supporting evidence) or by matching bias (a tendency to ignore negation and select items that lexically match elements of the rule). Results show that, like humans, LLMs reason better with deontic rules and display matching-bias-like errors. Together, these findings suggest that the performance of LLMs varies systematically across rule types and that their error patterns can parallel well-known human biases in this paradigm. --- ### A Reference Architecture of Reinforcement Learning Frameworks Authors: Xiaoran Liu, Istvan David Published: 2026-03-06 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.06413v1 The surge in reinforcement learning (RL) applications gave rise to diverse supporting technology, such as RL frameworks. However, the architectural patterns of these frameworks are inconsistent across implementations and there exists no reference architecture (RA) to form a common basis of comparison, evaluation, and integration. To address this gap, we propose an RA of RL frameworks. Through a grounded theory approach, we analyze 18 state-of-the-practice RL frameworks and, by that, we identify recurring architectural components and their relationships, and codify them in an RA. To demonstrate our RA, we reconstruct characteristic RL patterns. Finally, we identify architectural trends, e.g., commonly used components, and outline paths to improving RL frameworks. --- ### Physical Simulator In-the-Loop Video Generation Authors: Lin Geng Foo, Mark He Huang, Alexandros Lattas, Stylianos Moschoglou, Thabo Beeler, Christian Theobalt Published: 2026-03-06 Categories: cs.CV, cs.AI, cs.GR arXiv: https://arxiv.org/abs/2603.06408v1 Recent advances in diffusion-based video generation have achieved remarkable visual realism but still struggle to obey basic physical laws such as gravity, inertia, and collision. Generated objects often move inconsistently across frames, exhibit implausible dynamics, or violate physical constraints, limiting the realism and reliability of AI-generated videos. We address this gap by introducing Physical Simulator In-the-loop Video Generation (PSIVG), a novel framework that integrates a physical simulator into the video diffusion process. Starting from a template video generated by a pre-trained diffusion model, PSIVG reconstructs the 4D scene and foreground object meshes, initializes them within a physical simulator, and generates physically consistent trajectories. These simulated trajectories are then used to guide the video generator toward spatio-temporally physically coherent motion. To further improve texture consistency during object movement, we propose a Test-Time Texture Consistency Optimization (TTCO) technique that adapts text and feature embeddings based on pixel correspondences from the simulator. Comprehensive experiments demonstrate that PSIVG produces videos that better adhere to real-world physics while preserving visual quality and diversity. Project Page: https://vcai.mpi-inf.mpg.de/projects/PSIVG/ --- ### Locating and Editing Figure-Ground Organization in Vision Transformers Authors: Stefan Arnold, René Gröbner Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06407v1 Vision Transformers must resolve figure-ground organization by choosing between completions driven by local geometric evidence and those favored by global organizational priors, giving rise to a characteristic perceptual ambiguity. We aim to locate where the canonical Gestalt prior convexity is realized within the internal components of BEiT. Using a controlled perceptual conflict based on synthetic shapes of darts, we systematically mask regions that equally admit either a concave completion or a convex completion. We show that BEiT reliably favors convex completion under this competition. Projecting internal activations into the model's discrete visual codebook space via logit attribution reveals that this preference is governed by identifiable functional units within transformer substructures. Specifically, we find that figure-ground organization is ambiguous through early and intermediate layers and resolves abruptly in later layers. By decomposing the direct effect of attention heads, we identify head L0H9 acting as an early seed, introducing a weak bias toward convexity. Downscaling this single attention head shifts the distributional mass of the perceptual conflict across a continuous decision boundary, allowing concave evidence to guide completion. --- ### Adapter-Augmented Bandits for Online Multi-Constrained Multi-Modal Inference Scheduling Authors: Xianzhi Zhang, Yue Xu, Yinlin Zhu, Di Wu, Yipeng Zhou, Miao Hu, Guocong Quan Published: 2026-03-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.06403v1 Multi-modal large language model (MLLM) inference scheduling enables strong response quality under practical and heterogeneous budgets, beyond what a homogeneous single-backend setting can offer. Yet online MLLM task scheduling is nontrivial, as requests vary sharply in modality composition and latent reasoning difficulty, while execution backends incur distinct, time-varying costs due to system jitter and network variation. These coupled uncertainties pose two core challenges: deriving semantically faithful yet scheduling-relevant multi-modal task representations, and making low-overhead online decisions over irreversible multi-dimensional budgets. Accordingly, we propose \emph{M-CMAB} (\underline{M}ulti-modal \underline{M}ulti-constraint \underline{C}ontextual \underline{M}ulti-\underline{A}rmed \underline{B}andit), a multi-adapter-enhanced MLLM inference scheduling framework with three components: (i) a CLS-attentive, frozen-backbone \emph{Predictor} that extracts compact task representations and updates only lightweight adapters for action-specific estimation; (ii) a primal-dual \emph{Constrainer} that maintains online Lagrange multipliers to enforce long-horizon constraints via per-round objectives; and (iii) a two-phase \emph{Scheduler} that balances exploration and exploitation under irreversible budgets. We establish a regret guarantee under multi-dimensional knapsack constraints. On a composite multimodal benchmark with heterogeneous backends, \emph{M-CMAB} consistently outperforms state-of-the-art baselines across budget regimes, achieving up to 14.18% higher reward and closely tracking an oracle-aided upper bound. Codes are available at https://anonymous.4open.science/r/M2CMAB/. --- ### U6G XL-MIMO Radiomap Prediction: Multi-Config Dataset and Beam Map Approach Authors: Xiaojie Li, Yu Han, Zhizheng Lu, Shi Jin, Chao-Kai Wen Published: 2026-03-06 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2603.06401v1 The upper 6 GHz (U6G) band with XL-MIMO is a key enabler for sixth-generation wireless systems, yet intelligent radiomap prediction for such systems remains challenging. Existing datasets support only small-scale arrays (up to 8x8) with predominantly isotropic antennas, far from the 1024-element directional arrays envisioned for 6G. Moreover, current methods encode array configurations as scalar parameters, forcing neural networks to extrapolate array-specific radiation patterns, which fails when predicting radiomaps for configurations absent from training data. To jointly address data scarcity and generalization limitations, this paper advances XL-MIMO radiomap prediction from three aspects. To overcome data limitations, we construct the first XL-MIMO radiomap dataset containing 78400 radiomaps across 800 urban scenes, five frequency bands (1.8-6.7 GHz), and nine array configurations up to 32x32 uniform planar arrays with directional elements. To enable systematic evaluation, we establish a comprehensive benchmark framework covering practical scenarios from coverage estimation without field measurements to generalization across unseen configurations and environments. To enable generalization to arbitrary beam configurations without retraining, we propose the beam map, a physics-informed spatial feature that analytically computes array-specific coverage patterns. By decoupling deterministic array radiation from data learned multipath propagation, beam maps shift generalization from neural network extrapolation to physics-based computation. Integrating beam maps into existing architectures reduces mean absolute error by up to 60.0% when generalizing to unseen configurations and up to 50.5% when transferring to unseen environments. The complete dataset and code are publicly available at https://lxj321.github.io/MulticonfigRadiomapDataset/. --- ### DiffInf: Influence-Guided Diffusion for Supervision Alignment in Facial Attribute Learning Authors: Basudha Pal, Zhaoyang Wang, Rama Chellappa Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06399v2 Facial attribute classification relies on large-scale annotated datasets in which many traits, such as age and expression, are inherently ambiguous and continuous but are discretized into categorical labels. Annotation inconsistencies arise from subjectivity and visual confounders such as pose, illumination, expression, and demographic variation, creating mismatch between images and assigned labels. These inconsistencies introduce supervision errors that impair representation learning and degrade downstream prediction. We introduce DiffInf, a self-influence--guided diffusion framework for mitigating annotation inconsistencies in facial attribute learning. We first train a baseline classifier and compute sample-wise self-influence scores using a practical first-order approximation to identify training instances that disproportionately destabilize optimization. Instead of discarding these influential samples, we apply targeted generative correction via a latent diffusion autoencoder to better align visual content with assigned labels while preserving identity and realism. To enable differentiable guidance during correction, we train a lightweight predictor of high-influence membership and use it as a surrogate influence regularizer. The edited samples replace the originals, yielding an influence-refined dataset of unchanged size. Across multi-class facial attribute classification, DiffInf consistently improves generalization compared with standard noisy-label training, robust optimization baselines, and influence-based filtering. Our results demonstrate that repairing influential annotation inconsistencies at the image level enhances downstream facial attribute classification without sacrificing distributional coverage. --- ### DiffInf: Influence-Guided Diffusion for Supervision Alignment in Facial Attribute Learning Authors: Basudha Pal, Rama Chellappa Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06399v1 Facial attribute classification relies on large-scale annotated datasets in which many traits, such as age and expression, are inherently ambiguous and continuous but are discretized into categorical labels. Annotation inconsistencies arise from subjectivity and visual confounders such as pose, illumination, expression, and demographic variation, creating mismatch between images and assigned labels. These inconsistencies introduce supervision errors that impair representation learning and degrade downstream prediction. We introduce DiffInf, a self-influence--guided diffusion framework for mitigating annotation inconsistencies in facial attribute learning. We first train a baseline classifier and compute sample-wise self-influence scores using a practical first-order approximation to identify training instances that disproportionately destabilize optimization. Instead of discarding these influential samples, we apply targeted generative correction via a latent diffusion autoencoder to better align visual content with assigned labels while preserving identity and realism. To enable differentiable guidance during correction, we train a lightweight predictor of high-influence membership and use it as a surrogate influence regularizer. The edited samples replace the originals, yielding an influence-refined dataset of unchanged size. Across multi-class facial attribute classification, DiffInf consistently improves generalization compared with standard noisy-label training, robust optimization baselines, and influence-based filtering. Our results demonstrate that repairing influential annotation inconsistencies at the image level enhances downstream facial attribute classification without sacrificing distributional coverage. --- ### Efficient, Property-Aligned Fan-Out Retrieval via RL-Compiled Diffusion Authors: Pengcheng Jiang, Judith Yue Li, Moonkyung Ryu, R. Lily Hu, Kun Su, Zhong Yi Wan, Liam Hebert, Hao Peng, Jiawei Han, Dima Kuzmin, Craig Boutilier Published: 2026-03-06 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2603.06397v1 Many modern retrieval problems are set-valued: given a broad intent, the system must return a collection of results that optimizes higher-order properties (e.g., diversity, coverage, complementarity, coherence) while remaining grounded with respect to a fixed database. Set-valued objectives are typically non-decomposable and are not captured by existing supervised (query, content) datasets which only prioritize top-1 retrieval. Consequently, fan-out retrieval is often employed to generate diverse subqueries to retrieve item sets. While reinforcement learning (RL) can optimize set-level objectives via interaction, deploying an RL-tuned LLM for fan-out retrieval is prohibitively expensive at inference time. Conversely, diffusion-based generative retrieval enables efficient single-pass fan-out in embedding space, but requires objective-aligned training targets. To address these issues, we propose R4T (Retrieve-for-Train), which uses RL once as an objective transducer in a three-step process: (i) train a fan-out LLM with composite set-level rewards, (ii) synthesize objective-consistent training pairs, and (iii) train a lightweight diffusion retriever to model the conditional distribution of set-valued outputs. Across large-scale fashion and music benchmarks consisting of curated item sets, we show that R4T improves retrieval quality relative to strong baselines while reducing query-time fan-out latency by an order of magnitude. --- ### Talk Freely, Execute Strictly: Schema-Gated Agentic AI for Flexible and Reproducible Scientific Workflows Authors: Joel Strickland, Arjun Vijeta, Chris Moores, Oliwia Bodek, Bogdan Nenchev, Thomas Whitehead, Charles Phillips, Karl Tassenberg, Gareth Conduit, Ben Pellegrini Published: 2026-03-06 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2603.06394v1 Large language models (LLMs) can now translate a researcher's plain-language goal into executable computation, yet scientific workflows demand determinism, provenance, and governance that are difficult to guarantee when an LLM decides what runs. Semi-structured interviews with 18 experts across 10 industrial R&D stakeholders surface 2 competing requirements--deterministic, constrained execution and conversational flexibility without workflow rigidity--together with boundary properties (human-in-the-loop control and transparency) that any resolution must satisfy. We propose schema-gated orchestration as the resolving principle: the schema becomes a mandatory execution boundary at the composed-workflow level, so that nothing runs unless the complete action--including cross-step dependencies--validates against a machine-checkable specification. We operationalize the 2 requirements as execution determinism (ED) and conversational flexibility (CF), and use these axes to review 20 systems spanning 5 architectural groups along a validation-scope spectrum. Scores are assigned via a multi-model protocol--15 independent sessions across 3 LLM families--yielding substantial-to-near-perfect inter-model agreement (Krippendorff a=0.80 for ED and a=0.98 for CF), demonstrating that multi-model LLM scoring can serve as a reusable alternative to human expert panels for architectural assessment. The resulting landscape reveals an empirical Pareto front--no reviewed system achieves both high flexibility and high determinism--but a convergence zone emerges between the generative and workflow-centric extremes. We argue that a schema-gated architecture, separating conversational from execution authority, is positioned to decouple this trade-off, and distill 3 operational principles--clarification-before-execution, constrained plan-act orchestration, and tool-to-workflow-level gating--to guide adoption. --- ### Solving Jigsaw Puzzles in the Wild: Human-Guided Reconstruction of Cultural Heritage Fragments Authors: Omidreza Safaei, Sinem Aslan, Sebastiano Vascon, Luca Palmieri, Marina Khoroshiltseva, Marcello Pelillo Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06389v1 Reassembling real-world archaeological artifacts from fragmented pieces poses significant challenges due to erosion, missing regions, irregular shapes, and large-scale ambiguity. Traditional jigsaw puzzle solvers, often designed for clean synthetic scenarios, struggle under these conditions, especially when the number of fragments grows into the thousands, as in the RePAIR benchmark. In this paper, we propose a human-in-the-loop (HIL) puzzle solving framework designed to address the complexity and scale of real-world cultural heritage reconstruction. Our approach integrates an automatic relaxation-labeling solver with interactive human guidance, allowing users to iteratively lock verified placements, correct errors, and guide the system toward semantically and geometrically coherent assemblies. We introduce two complementary interaction strategies, Iterative Anchoring and Continuous Interactive Refinement, which support scalable reconstruction across varying levels of ambiguity and puzzle size. Experiments on several RePAIR groups demonstrate that our hybrid approach substantially outperforms both fully automatic and manual baselines in accuracy and efficiency, offering a practical solution for large-scale expert-in-the-loop artifact reassembly. --- ### REACT++: Efficient Cross-Attention for Real-Time Scene Graph Generation Authors: Maëlic Neau, Zoe Falomir Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06386v1 Scene Graph Generation (SGG) is a task that encodes visual relationships between objects in images as graph structures. SGG shows significant promise as a foundational component for downstream tasks, such as reasoning for embodied agents. To enable real-time applications, SGG must address the trade-off between performance and inference speed. However, current methods tend to focus on one of the following: (1) improving relation prediction accuracy, (2) enhancing object detection accuracy, or (3) reducing latency, without aiming to balance all three objectives simultaneously. To address this limitation, we build on the powerful Real-time Efficiency and Accuracy Compromise for Tradeoffs in Scene Graph Generation (REACT) architecture and propose REACT++, a new state-of-the-art model for real-time SGG. By leveraging efficient feature extraction and subject-to-object cross-attention within the prototype space, REACT++ balances latency and representational power. REACT++ achieves the highest inference speed among existing SGG models, improving relation prediction accuracy without sacrificing object detection performance. Compared to the previous REACT version, REACT++ is 20% faster with a gain of 10% in relation prediction accuracy on average. The code is available at https://github.com/Maelic/SGG-Benchmark. --- ### Prompt Group-Aware Training for Robust Text-Guided Nuclei Segmentation Authors: Yonghuang Wu, Zhenyang Liang, Wenwen Zeng, Xuan Xie, Jinhua Yu Published: 2026-03-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.06384v1 Foundation models such as Segment Anything Model 3 (SAM3) enable flexible text-guided medical image segmentation, yet their predictions remain highly sensitive to prompt formulation. Even semantically equivalent descriptions can yield inconsistent masks, limiting reliability in clinical and pathology workflows. We reformulate prompt sensitivity as a group-wise consistency problem. Semantically related prompts are organized into \emph{prompt groups} sharing the same ground-truth mask, and a prompt group-aware training framework is introduced for robust text-guided nuclei segmentation. The approach combines (i) a quality-guided group regularization that leverages segmentation loss as an implicit ranking signal, and (ii) a logit-level consistency constraint with a stop-gradient strategy to align predictions within each group. The method requires no architectural modification and leaves inference unchanged. Extensive experiments on multi-dataset nuclei benchmarks show consistent gains under textual prompting and markedly reduced performance variance across prompt quality levels. On six zero-shot cross-dataset tasks, our method improves Dice by an average of 2.16 points. These results demonstrate improved robustness and generalization for vision-language segmentation in computational pathology. --- ### CHMv2: Improvements in Global Canopy Height Mapping using DINOv3 Authors: John Brandt, Seungeun Yi, Jamie Tolan, Xinyuan Li, Peter Potapov, Jessica Ertel, Justine Spore, Huy V. Vo, Michaël Ramamonjisoa, Patrick Labatut, Piotr Bojanowski, Camille Couprie Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06382v1 Accurate canopy height information is essential for quantifying forest carbon, monitoring restoration and degradation, and assessing habitat structure, yet high-fidelity measurements from airborne laser scanning (ALS) remain unevenly available globally. Here we present CHMv2, a global, meter-resolution canopy height map derived from high-resolution optical satellite imagery using a depth-estimation model built on DINOv3 and trained against ALS canopy height models. Compared to existing products, CHMv2 substantially improves accuracy, reduces bias in tall forests, and better preserves fine-scale structure such as canopy edges and gaps. These gains are enabled by a large expansion of geographically diverse training data, automated data curation and registration, and a loss formulation and data sampling strategy tailored to canopy height distributions. We validate CHMv2 against independent ALS test sets and against tens of millions of GEDI and ICESat-2 observations, demonstrating consistent performance across major forest biomes. --- ### Kinetic-based regularization: Learning spatial derivatives and PDE applications Authors: Abhisek Ganguly, Santosh Ansumali, Sauro Succi Published: 2026-03-06 Categories: math.NA, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.06380v1 Accurate estimation of spatial derivatives from discrete and noisy data is central to scientific machine learning and numerical solutions of PDEs. We extend kinetic-based regularization (KBR), a localized multidimensional kernel regression method with a single trainable parameter, to learn spatial derivatives with provable second-order accuracy in 1D. Two derivative-learning schemes are proposed: an explicit scheme based on the closed-form prediction expressions, and an implicit scheme that solves a perturbed linear system at the points of interest. The fully localized formulation enables efficient, noise-adaptive derivative estimation without requiring global system solving or heuristic smoothing. Both approaches exhibit quadratic convergence, matching second-order finite difference for clean data, along with a possible high-dimensional formulation. Preliminary results show that coupling KBR with conservative solvers enables stable shock capture in 1D hyperbolic PDEs, acting as a step towards solving PDEs on irregular point clouds in higher dimensions while preserving conservation laws. --- ### MoEMambaMIL: Structure-Aware Selective State Space Modeling for Whole-Slide Image Analysis Authors: Dongqing Xie, Yonghuang Wu Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06378v1 Whole-slide image (WSI) analysis is challenging due to the gigapixel scale of slides and their inherent hierarchical multi-resolution structure. Existing multiple instance learning (MIL) approaches often model WSIs as unordered collections of patches, which limits their ability to capture structured dependencies between global tissue organization and local cellular patterns. Although recent State Space Models (SSMs) enable efficient modeling of long sequences, how to structure WSI tokens to fully exploit their spatial hierarchy remains an open problem.We propose MoEMambaMIL, a structure-aware SSM framework for WSI analysis that integrates region-nested selective scanning with mixture-of-experts (MoE) modeling. Leveraging multi-resolution preprocessing, MoEMambaMIL organizes patch tokens into region-aware sequences that preserve spatial containment across resolutions. On top of this structured sequence, we decouple resolution-aware encoding and region-adaptive contextual modeling via a combination of static, resolution-specific experts and dynamic sparse experts with learned routing. This design enables efficient long-sequence modeling while promoting expert specialization across heterogeneous diagnostic patterns. Experiments demonstrate that MoEMambaMIL achieves the best performance across 9 downstream tasks. --- ### Rewis3d: Reconstruction Improves Weakly-Supervised Semantic Segmentation Authors: Jonas Ernst, Wolfgang Boettcher, Lukas Hoyer, Jan Eric Lenssen, Bernt Schiele Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06374v1 We present Rewis3d, a framework that leverages recent advances in feed-forward 3D reconstruction to significantly improve weakly supervised semantic segmentation on 2D images. Obtaining dense, pixel-level annotations remains a costly bottleneck for training segmentation models. Alleviating this issue, sparse annotations offer an efficient weakly-supervised alternative. However, they still incur a performance gap. To address this, we introduce a novel approach that leverages 3D scene reconstruction as an auxiliary supervisory signal. Our key insight is that 3D geometric structure recovered from 2D videos provides strong cues that can propagate sparse annotations across entire scenes. Specifically, a dual student-teacher architecture enforces semantic consistency between 2D images and reconstructed 3D point clouds, using state-of-the-art feed-forward reconstruction to generate reliable geometric supervision. Extensive experiments demonstrate that Rewis3d achieves state-of-the-art performance in sparse supervision, outperforming existing approaches by 2-7% without requiring additional labels or inference overhead. --- ### Adaptive Lipschitz-Free Conditional Gradient Methods for Stochastic Composite Nonconvex Optimization Authors: Ganzhao Yuan Published: 2026-03-06 Categories: cs.LG, math.NA, math.OC arXiv: https://arxiv.org/abs/2603.06369v1 We propose ALFCG (Adaptive Lipschitz-Free Conditional Gradient), the first \textit{adaptive} projection-free framework for stochastic composite nonconvex minimization that \textit{requires neither global smoothness constants nor line search}. Unlike prior conditional gradient methods that use openloop diminishing stepsizes, conservative Lipschitz constants, or costly backtracking, ALFCG maintains a self-normalized accumulator of historical iterate differences to estimate local smoothness and minimize a quadratic surrogate model at each step. This retains the simplicity of Frank-Wolfe while adapting to unknown geometry. We study three variants. ALFCG-FS addresses finite-sum problems with a SPIDER estimator. ALFCG-MVR1 and ALFCG-MVR2 handle stochastic expectation problems by using momentum-based variance reduction with single-batch and two-batch updates, and operate under average and individual smoothness, respectively. To reach an $ε$-stationary point, ALFCG-FS attains $\mathcal{O}(N+\sqrt{N}ε^{-2})$ iteration complexity, while ALFCG-MVR1 and ALFCG-MVR2 achieve $\tilde{\mathcal{O}}(σ^2ε^{-4}+ε^{-2})$ and $\tilde{\mathcal{O}}(σε^{-3}+ε^{-2})$, where $N$ is the number of components and $σ$ is the noise level. In contrast to typical $\mathcal{O}(ε^{-4})$ or $\mathcal{O}(ε^{-3})$ rates, our bounds reduce to the optimal rate up to logarithmic factors $\tilde{\mathcal{O}}(ε^{-2})$ as the noise level $σ\to 0$. Extensive experiments on multiclass classification over nuclear norm balls and $\ell_p$ balls show that ALFCG generally outperforms state-of-the-art conditional gradient baselines. --- ### OralGPT-Plus: Learning to Use Visual Tools via Reinforcement Learning for Panoramic X-ray Analysis Authors: Yuxuan Fan, Jing Hao, Hong Chen, Jiahao Bao, Yihua Shao, Yuci Liang, Kuo Feng Hung, Hao Tang Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06366v1 Panoramic dental radiographs require fine-grained spatial reasoning, bilateral symmetry understanding, and multi-step diagnostic verification, yet existing vision-language models operate under a static single-pass paradigm that limits their clinical reliability. In this paper, we introduce OralGPT-Plus, an agentic vision-language model designed to perform iterative and symmetry-aware diagnostic reasoning for panoramic dental radiograph analysis. To support this paradigm, we construct DentalProbe, a five-thousand-image dataset with expert-curated diagnostic trajectories that provide structured supervision for localized inspection and contralateral comparison. We further develop a Reinspection-driven reinforcement learning framework that encourages clinically meaningful re-examination and stabilizes long-horizon reasoning with rubric-based reward and conditioned diagnostic-driven reward. In parallel, we present MMOral-X, the first benchmark for holistic panoramic diagnosis, containing 300 open-ended questions and region-level annotations across multiple difficulty levels. OralGPT-Plus demonstrates consistent and reliable improvements over strong baselines on MMOral-X and established panoramic benchmarks, indicating the effectiveness of interactive and symmetry-informed reasoning. Our work highlights the value of agentic modeling for dental imaging and provides a foundation for future research in clinically aligned panoramic radiograph analysis. --- ### ESAA-Security: An Event-Sourced, Verifiable Architecture for Agent-Assisted Security Audits of AI-Generated Code Authors: Elzo Brito dos Santos Filho Published: 2026-03-06 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2603.06365v1 AI-assisted software generation has increased development speed, but it has also amplified a persistent engineering problem: systems that are functionally correct may still be structurally insecure. In practice, prompt-based security review with large language models often suffers from uneven coverage, weak reproducibility, unsupported findings, and the absence of an immutable audit trail. The ESAA architecture addresses a related governance problem in agentic software engineering by separating heuristic agent cognition from deterministic state mutation through append-only events, constrained outputs, and replay-based verification. This paper presents ESAA-Security, a domain-specific specialization of ESAA for agent-assisted security auditing of software repositories, with particular emphasis on AI-generated or AI-modified code. ESAA-Security structures auditing as a governed execution pipeline with four phases reconnaissance, domain audit execution, risk classification, and final reporting and operationalizes the workflow into 26 tasks, 16 security domains, and 95 executable checks. The framework produces structured check results, vulnerability inventories, severity classifications, risk matrices, remediation guidance, executive summaries, and a final markdown/JSON audit report. The central idea is that security review should not be modeled as a free-form conversation with an LLM, but as an evidence-oriented audit process governed by contracts and events. In ESAA-Security, agents emit structured intentions under constrained protocols; the orchestrator validates them, persists accepted outputs to an append-only log, reprojects derived views, and verifies consistency through replay and hashing. The result is a traceable, reproducible, and risk-oriented audit architecture whose final report is auditable by construction. --- ### Computer vision-based estimation of invertebrate biomass Authors: Mikko Impiö, Philipp M. Rehsen, Jarrett Blair, Cecilie Mielec, Arne J. Beermann, Florian Leese, Toke T. Høye, Jenni Raitoharju Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06362v1 The ability to estimate invertebrate biomass using only images could help scaling up quantitative biodiversity monitoring efforts. Computer vision-based methods have the potential to omit the manual, time-consuming, and destructive process of dry weighing specimens. We present two approaches for dry mass estimation that do not require additional manual effort apart from imaging the specimens: fitting a linear model with novel predictors, automatically calculated by an imaging device, and training a family of end-to-end deep neural networks for the task, using single-view, multi-view, and metadata-aware architectures. We propose using area and sinking speed as predictors. These can be calculated with BIODISCOVER, which is a dual-camera system that captures image sequences of specimens sinking in an ethanol column. For this study, we collected a large dataset of dry mass measurement and image sequence pairs to train and evaluate models. We show that our methods can estimate specimen dry mass even with complex and visually diverse specimen morphologies. Combined with automatic taxonomic classification, our approach is an accurate method for group-level dry mass estimation, with a median percentage error of 10-20% for individuals. We highlight the importance of choosing appropriate evaluation metrics, and encourage using both percentage errors and absolute errors as metrics, because they measure different properties. We also explore different optimization losses, data augmentation methods, and model architectures for training deep-learning models. --- ### CLAIRE: Compressed Latent Autoencoder for Industrial Representation and Evaluation -- A Deep Learning Framework for Smart Manufacturing Authors: Mohammadhossein Ghahramani, Mengchu Zhou Published: 2026-03-06 Categories: cs.LG, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2603.06361v1 Accurate fault detection in high-dimensional industrial environments remains a major challenge due to the inherent complexity, noise, and redundancy in sensor data. This paper introduces CLAIRE, i.e., a hybrid end-to-end learning framework that integrates unsupervised deep representation learning with supervised classification for intelligent quality control in smart manufacturing systems. It employs an optimized deep autoencoder to transform raw input into a compact latent space, effectively capturing the intrinsic data structure while suppressing irrelevant or noisy features. The learned representations are then fed into a downstream classifier to perform binary fault prediction. Experimental results on a high-dimensional dataset demonstrate that CLAIRE significantly outperforms conventional classifiers trained directly on raw features. Moreover, the framework incorporates a post hoc phase, using a game-theory-based interpretability technique, to analyze the latent space and identify the most informative input features contributing to fault predictions. The proposed framework highlights the potential of integrating explainable AI with feature-aware regularization for robust fault detection. The modular and interpretable nature of the proposed framework makes it highly adaptable, offering promising applications in other domains characterized by complex, high-dimensional data, such as healthcare, finance, and environmental monitoring. --- ### Tiny, Hardware-Independent, Compression-based Classification Authors: Charles Meyers, Aaron MacSween, Erik Elmroth, Tommy Löfstedt Published: 2026-03-06 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2603.06359v1 The recent developments in machine learning have highlighted a conflict between online platforms and their users in terms of privacy. The importance of user privacy and the struggle for power over user data has been intensified as regulators and operators attempt to police online platforms. As users have become increasingly aware of privacy issues, client-side data storage, management, and analysis have become a favoured approach to large-scale centralised machine learning. However, state-of-the-art machine learning methods require vast amounts of labelled user data, making them unsuitable for models that reside client-side and only have access to a single user's data. State-of-the-art methods are also computationally expensive, which degrades the user experience on compute-limited hardware and also reduces battery life. A recent alternative approach has proven remarkably successful in classification tasks across a wide variety of data -- using a compression-based distance measure (called normalised compression distance) to measure the distance between generic objects in classical distance-based machine learning methods. In this work, we demonstrate that the normalised compression distance is actually not a metric; develop it for the wider context of kernel methods to allow modelling of complex data; and present techniques to improve the training time of models that use this distance measure. We demonstrate that the normalised compression distance works as well as and sometimes better than other metrics and kernels -- while requiring only marginally more computational costs and in spite of the lack of formal metric properties. The end results is a simple model with remarkable accuracy even when trained on a very small number of samples allowing for models that are small and effective enough to run entirely on a client device using only user-supplied data. --- ### LATO: 3D Mesh Flow Matching with Structured TOpology Preserving LAtents Authors: Tianhao Zhao, Youjia Zhang, Hang Long, Jinshen Zhang, Wenbing Li, Yang Yang, Gongbo Zhang, Jozef Hladký, Matthias Nießner, Wei Yang Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06357v1 In this paper, we introduce LATO, a novel topology-preserving latent representation that enables scalable, flow matching-based synthesis of explicit 3D meshes. LATO represents a mesh as a Vertex Displacement Field (VDF) anchored on surface, incorporating a sparse voxel Variational Autoencoder (VAE) to compress this explicit signal into a structured, topology-aware voxel latent. To decapsulate the mesh, the VAE decoder progressively subdivides and prunes latent voxels to instantiate precise vertex locations. In the end, a dedicated connection head queries the voxel latent to predict edge connectivity between vertex pairs directly, allowing mesh topology to be recovered without isosurface extraction or heuristic meshing. For generative modeling, LATO adopts a two-stage flow matching process, first synthesizing the structure voxels and subsequently refining the voxel-wise topology features. Compared to prior isosurface/triangle-based diffusion models and autoregressive generation approaches, LATO generates meshes with complex geometry, well-formed topology while being highly efficient in inference. --- ### Frequency-Separable Hamiltonian Neural Network for Multi-Timescale Dynamics Authors: Yaojun Li, Yulong Yang, Christine Allen-Blanchette Published: 2026-03-06 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2603.06354v2 While Hamiltonian mechanics provides a powerful inductive bias for neural networks modeling dynamical systems, Hamiltonian Neural Networks and their variants often fail to capture complex temporal dynamics spanning multiple timescales. This limitation is commonly linked to the spectral bias of deep neural networks, which favors learning low-frequency, slow-varying dynamics. Prior approaches have sought to address this issue through symplectic integration schemes that enforce energy conservation or by incorporating geometric constraints to impose structure on the configuration-space. However, such methods either remain limited in their ability to fully capture multiscale dynamics or require substantial domain specific assumptions. In this work, we exploit the observation that Hamiltonian functions admit decompositions into explicit fast and slow modes and can be reconstructed from these components. We introduce the Frequency-Separable Hamiltonian Neural Network (FS-HNN), which parameterizes the system Hamiltonian using multiple networks, each governed by Hamiltonian dynamics and trained on data sampled at distinct timescales. We further extend this framework to partial differential equations by learning a state- and boundary-conditioned symplectic operators. Empirically, we show that FS-HNN improves long-horizon extrapolation performance on challenging dynamical systems and generalizes across a broad range of ODE and PDE problems. --- ### DC-DiT: Adaptive Compute and Elastic Inference for Visual Generation via Dynamic Chunking Authors: Akash Haridas, Utkarsh Saxena, Parsa Ashrafi Fashi, Mehdi Rezagholizadeh, Vikram Appia, Emad Barsoum Published: 2026-03-06 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.06351v2 Diffusion Transformers rely on static patchify tokenization, assigning the same token budget to smooth backgrounds, detailed object regions, noisy early timesteps, and late-stage refinements. We introduce the Dynamic Chunking Diffusion Transformer (DC-DiT), which replaces fixed patchification with a learned encoder-router-decoder scaffold that adaptively compresses the 2D input into a shorter token sequence through a chunking mechanism learned end-to-end with diffusion training. DC-DiT allocates fewer tokens to predictable regions and noisy timesteps, and more tokens to detailed regions and later refinement stages, yielding meaningful spatial segmentations and timestep-adaptive compression schedules without supervision. Furthermore, the router provides an importance ordering over retained tokens, enabling elastic inference: a single checkpoint can be evaluated at flexible compute budgets with a smooth quality-compute tradeoff. Additionally, DC-DiT can be upcycled from pretrained DiT checkpoints and is also compatible with orthogonal dynamic computation approaches. On class-conditional ImageNet generation, DC-DiT reduces inference FLOPs by up to 36.8% and improves FID by up to 37.8% over DiT baselines, yielding a stronger quality--compute Pareto frontier across model scales, resolutions, and guidance settings. More broadly, these results suggest that adaptive tokenization is a general mechanism for making visual generation both more efficient and more flexible at inference time. --- ### MoEless: Efficient MoE LLM Serving via Serverless Computing Authors: Hanfei Yu, Bei Ouyang, Shwai He, Ang Li, Hao Wang Published: 2026-03-06 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.06350v1 Large Language Models (LLMs) have become a cornerstone of AI, driving progress across diverse domains such as content creation, search and recommendation systems, and AI-assisted workflows. To alleviate extreme training costs and advancing model scales, Mixture-of-Experts (MoE) has become a popular backbone for modern LLMs, which are commonly served in distributed deployment using expert parallelism (EP). However, MoE's sparse activation mechanism leads to severe expert load imbalance, where a few experts become overloaded while others remain idle, resulting in expert stragglers that inflate inference latency and serving cost. Existing expert load balancing solutions assume static resource configurations on serverful infrastructures, limiting expert scalability and elasticity, and resulting in either costly real-time expert swapping or degraded generation quality. We present MoEless, the first serverless MoE serving framework that mitigates expert load imbalance and accelerates inference via serverless experts. MoEless employs lightweight, layer-aware predictors to accurately estimate incoming expert load distributions and proactively identify stragglers. We design optimized expert scaling and placement strategies to maximize function locality, improve GPU utilization, and balance loads across experts and GPUs. MoEless is prototyped on top of Megatron-LM and deployed on an eight-GPU testbed. Experiments with open-source MoE models and real-world workloads show that MoEless reduces inference latency by 43% and inference cost by 84% compared to state-of-the-art solutions. --- ### Transparent AI for Mathematics: Transformer-Based Large Language Models for Mathematical Entity Relationship Extraction with XAI Authors: Tanjim Taharat Aurpa Published: 2026-03-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.06348v1 Mathematical text understanding is a challenging task due to the presence of specialized entities and complex relationships between them. This study formulates mathematical problem interpretation as a Mathematical Entity Relation Extraction (MERE) task, where operands are treated as entities and operators as their relationships. Transformer-based models are applied to automatically extract these relations from mathematical text, with Bidirectional Encoder Representations from Transformers (BERT) achieving the best performance, reaching an accuracy of 99.39%. To enhance transparency and trust in the model's predictions, Explainable Artificial Intelligence (XAI) is incorporated using Shapley Additive Explanations (SHAP). The explainability analysis reveals how specific textual and mathematical features influence relation prediction, providing insights into feature importance and model behavior. By combining transformer-based learning, a task-specific dataset, and explainable modeling, this work offers an effective and interpretable framework for MERE, supporting future applications in automated problem solving, knowledge graph construction, and intelligent educational systems. --- ### K-MaT: Knowledge-Anchored Manifold Transport for Cross-Modal Prompt Learning in Medical Imaging Authors: Jiajun Zeng, Shadi Albarqouni Published: 2026-03-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.06340v1 Large-scale biomedical vision-language models (VLMs) adapted on high-end imaging (e.g., CT) often fail to transfer to frontline low-end modalities (e.g., radiography), collapsing into modality-specific shortcuts. We propose K-MaT (Knowledge-Anchored Manifold Transport), a prompt-learning framework that transfers decision structures to low-end modalities without requiring low-end training images. K-MaT factorizes prompts, anchors them to clinical text descriptions, and aligns the low-end prompt manifold to the visually-grounded high-end space using Fused Gromov-Wasserstein optimal transport. We evaluate K-MaT on four cross-modal benchmarks, including dermoscopy, mammography to ultrasound, and CT to chest X-ray. K-MaT achieves state-of-the-art results, improving the average harmonic mean of accuracy to 44.1% (from BiomedCoOp's 42.0%) and macro-F1 to 36.2%. Notably, on the challenging breast imaging task, it mitigates the catastrophic forgetting seen in standard methods like CoOp (which drops to 27.0% accuracy on the low-end), preserving robust performance across modalities. Aligning prompt manifolds via optimal transport provides a highly effective route for the zero-shot cross-modal deployment of medical VLMs. --- ### AI End-to-End Radiation Treatment Planning Under One Second Authors: Simon Arberet, Riqiang Gao, Martin Kraus, Florin C. Ghesu, Wilko Verbakel, Mamadou Diallo, Anthony Magliari, Venkatesan Karuppusamy, Sushil Beriwal, REQUITE Consortium, Ali Kamen, Dorin Comaniciu Published: 2026-03-06 Categories: eess.IV, cs.AI, cs.LG, eess.SY, physics.med-ph arXiv: https://arxiv.org/abs/2603.06338v1 Artificial intelligence-based radiation therapy (RT) planning has the potential to reduce planning time and inter-planner variability, improving efficiency and consistency in clinical workflows. Most existing automated approaches rely on multiple dose evaluations and corrections, resulting in plan generation times of several minutes. We introduce AIRT (Artificial Intelligence-based Radiotherapy), an end-to-end deep-learning framework that directly infers deliverable treatment plans from CT images and structure contours. AIRT generates single-arc VMAT prostate plans, from imaging and anatomical inputs to leaf sequencing, in under one second on a single Nvidia A100 GPU. The framework includes a differentiable dose feedback, an adversarial fluence map shaping, and a plan generation augmentation to improve plan quality and robustness. The model was trained on more than 10,000 intact prostate cases. Non-inferiority to RapidPlan Eclipse was demonstrated across target coverage and OAR sparing metrics. Target homogeneity (HI = 0.10 $\pm$ 0.01) and OAR sparing were similar to reference plans when evaluated using AcurosXB. These results represent a significant step toward ultra-fast standardized RT planning and a streamlined clinical workflow. --- ### SAHOO: Safeguarded Alignment for High-Order Optimization Objectives in Recursive Self-Improvement Authors: Subramanyam Sahoo, Aman Chadha, Vinija Jain, Divya Chaudhary Published: 2026-03-06 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.06333v1 Recursive self-improvement is moving from theory to practice: modern systems can critique, revise, and evaluate their own outputs, yet iterative self-modification risks subtle alignment drift. We introduce SAHOO, a practical framework to monitor and control drift through three safeguards: (i) the Goal Drift Index (GDI), a learned multi-signal detector combining semantic, lexical, structural, and distributional measures; (ii) constraint preservation checks that enforce safety-critical invariants such as syntactic correctness and non-hallucination; and (iii) regression-risk quantification to flag improvement cycles that undo prior gains. Across 189 tasks in code generation, mathematical reasoning, and truthfulness, SAHOO produces substantial quality gains, including 18.3 percent improvement in code tasks and 16.8 percent in reasoning, while preserving constraints in two domains and maintaining low violations in truthfulness. Thresholds are calibrated on a small validation set of 18 tasks across three cycles. We further map the capability-alignment frontier, showing efficient early improvement cycles but rising alignment costs later and exposing domain-specific tensions such as fluency versus factuality. SAHOO therefore makes alignment preservation during recursive self-improvement measurable, deployable, and systematically validated at scale. --- ### WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching Authors: Weilun Feng, Guoxin Fan, Haotong Qin, Mingqiang Wu, Yuqi Li, Xiangqi Li, Zhulin An, Libo Huang, Dingrui Wang, Longlong Liao, Michele Magno, Yongjun Xu et al. Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06331v2 Diffusion-based world models have shown strong potential for unified world simulation, but the iterative denoising remains too costly for interactive use and long-horizon rollouts. While feature caching can accelerate inference without training, we find that policies designed for single-modal diffusion transfer poorly to world models due to two world-model-specific obstacles: \emph{token heterogeneity} from multi-modal coupling and spatial variation, and \emph{non-uniform temporal dynamics} where a small set of hard tokens drives error growth, making uniform skipping either unstable or overly conservative. We propose \textbf{WorldCache}, a caching framework tailored to diffusion world models. We introduce \textit{Curvature-guided Heterogeneous Token Prediction}, which uses a physics-grounded curvature score to estimate token predictability and applies a Hermite-guided damped predictor for chaotic tokens with abrupt direction changes. We also design \textit{Chaotic-prioritized Adaptive Skipping}, which accumulates a curvature-normalized, dimensionless drift signal and recomputes only when bottleneck tokens begin to drift. Experiments on diffusion world models show that WorldCache delivers up to \textbf{3.7$\times$} end-to-end speedups while maintaining \textbf{98\%} rollout quality, demonstrating the vast advantages and practicality of WorldCache in resource-constrained scenarios. Our code is released in https://github.com/FofGofx/WorldCache. --- ### Structured Exploration vs. Generative Flexibility: A Field Study Comparing Bandit and LLM Architectures for Personalised Health Behaviour Interventions Authors: Dominik P. Hofer, Haochen Song, Rania Islambouli, Laura Hawkins, Ananya Bhattacharjee, Meredith Franklin, Joseph Jay Williams, Jan D. Smeddinck Published: 2026-03-06 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.06330v1 Behaviour Change Techniques (BCTs) are central to digital health interventions, yet selecting and delivering effective techniques remains challenging. Contextual bandits enable statistically grounded optimisation of BCT selection, while Large Language Models (LLMs) offer flexible, context-sensitive message generation. We conducted a 4-week study on physical activity motivation (N=54; 9 post-study interviews) that compared five daily messaging approaches: random templates, contextual bandit with templates, LLM generation, hybrid bandit+LLM, and LLM with interaction history. LLM-based approaches were rated substantially more helpful than templates, but no significant differences emerged among LLM conditions. Unexpectedly, bandit optimisation for BCTs selection yielded no additional perceived helpfulness compared with LLM-only approaches. Unconstrained LLMs focused heavily on a single BCT, whereas bandit systems enforced systematic exploration-exploitation across techniques. Quantitative and qualitative findings suggest contextual acknowledgement of user input drove perceived helpfulness. We contribute design suggestions for reflective AI health behaviour change systems that address a trade-off between structured exploration and generative autonomy. --- ### The Art That Poses Back: Assessing AI Pastiches after Contemporary Artworks Authors: Anca Dinu, Andreiana Mihail, Andra-Maria Florescu, Claudiu Creanga Published: 2026-03-06 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2603.06324v1 This study explores artificial visual creativity, focusing on ChatGPT's ability to generate new images intentionally pastiching original artworks such as paintings, drawings, sculptures and installations. The process involved twelve artists from Romania, Bulgaria, France, Austria, and the United Kingdom, each invited to contribute with three of their artworks and to grade and comment on the AI-generated versions. The analysis combines human evaluation with computational methods aimed at detecting visual and stylistic similarities or divergences between the original works and their AI-produced renditions. The results point to a significant gap between color and texture-based similarity and compositional, conceptual, and perceptual one. Consequently, we advocate for the use of a "style transfer dashboard" of complementary metrics to evaluate the similarity between pastiches and originals, rather than using a single style metric. The artists' comments revealed limitations of ChatGPT's pastiches after contemporary artworks, which were perceived by the authors of the originals as lacking dimensionality, context, and intentional sense, and seeming more of a paraphrase or an approximate quotation rather than as a valuable, emotion-evoking artwork. --- ### P-SLCR: Unsupervised Point Cloud Semantic Segmentation via Prototypes Structure Learning and Consistent Reasoning Authors: Lixin Zhan, Jie Jiang, Tianjian Zhou, Yukun Du, Yan Zheng, Xuehu Duan Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06321v1 Current semantic segmentation approaches for point cloud scenes heavily rely on manual labeling, while research on unsupervised semantic segmentation methods specifically for raw point clouds is still in its early stages. Unsupervised point cloud learning poses significant challenges due to the absence of annotation information and the lack of pre-training. The development of effective strategies is crucial in this context. In this paper, we propose a novel prototype library-driven unsupervised point cloud semantic segmentation strategy that utilizes Structure Learning and Consistent Reasoning (P-SLCR). First, we propose a Consistent Structure Learning to establish structural feature learning between consistent points and the library of consistent prototypes by selecting high-quality features. Second, we propose a Semantic Relation Consistent Reasoning that constructs a prototype inter-relation matrix between consistent and ambiguous prototype libraries separately. This process ensures the preservation of semantic consistency by imposing constraints on consistent and ambiguous prototype libraries through the prototype inter-relation matrix. Finally, our method was extensively evaluated on the S3DIS, SemanticKITTI, and Scannet datasets, achieving the best performance compared to unsupervised methods. Specifically, the mIoU of 47.1% is achieved for Area-5 of the S3DIS dataset, surpassing the classical fully supervised method PointNet by 2.5%. ---