# Research Papers — 2025-12 · bundle page 141 of 175 > Full-text markdown of papers #7001–7050 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2025-12/page/141 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/140.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/142.md) ### Teaching Language Models Mechanistic Explainability Through Arrow-Pushing Authors: Théo A. Neukomm, Zlatko Jončev, Philippe Schwaller Published: 2025-12-05 Categories: cs.LG, physics.chem-ph arXiv: https://arxiv.org/abs/2512.05722v1 Chemical reaction mechanisms provide crucial insight into synthesizability, yet current Computer-Assisted Synthesis Planning (CASP) systems lack mechanistic grounding. We introduce a computational framework for teaching language models to predict chemical reaction mechanisms through arrow pushing formalism, a century-old notation that tracks electron flow while respecting conservation laws. We developed MechSMILES, a compact textual format encoding molecular structure and electron flow, and trained language models on four mechanism prediction tasks of increasing complexity using mechanistic reaction datasets, such as mech-USPTO-31k and FlowER. Our models achieve more than 95\% top-3 accuracy on elementary step prediction and scores that surpass 73\% on mech-USPTO-31k, and 93\% on FlowER dataset for the retrieval of complete reaction mechanisms on our hardest task. This mechanistic understanding enables three key applications. First, our models serve as post-hoc validators for CASP systems, filtering chemically implausible transformations. Second, they enable holistic atom-to-atom mapping that tracks all atoms, including hydrogens. Third, they extract catalyst-aware reaction templates that distinguish recycled catalysts from spectator species. By grounding predictions in physically meaningful electron moves that ensure conservation of mass and charge, this work provides a pathway toward more explainable and chemically valid computational synthesis planning, while providing an architecture-agnostic framework for the benchmarking of mechanism prediction. --- ### BERTO: Intent-Driven Network Time Series Forecasting via Natural Language Operator Preferences Authors: Nitin Priyadarshini Shankar, Vaibhav Singh, Sheetal Kalyani, Christian Maciocco Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05721v2 Traditional cellular traffic forecasting models are optimized for minimizing symmetric errors, leaving them indifferent to shifting operational priorities. To bridge this gap, we introduce BERTO, a BERT-based framework for traffic prediction and energy optimization in cellular networks. Built on transformer architectures, BERTO achieves high prediction accuracy while enabling a single fine-tuned model to operate across multiple forecasting regimes via natural-language operator prompts. By combining a Balancing Loss Function (BLF) with prompt-based conditioning, BERTO adaptively shifts its forecasting bias toward underprediction or overprediction depending on the operator's desired trade-off between power savings and service quality. This allows the same model to dynamically generate different decision-aware forecasts without retraining or modifying model parameters. Experiments on real-world datasets demonstrate that BERTO can operate across a flexible range of approximately 1.4 kW in power consumption while balancing 9x variation in service level agreement (SLA) violations, making it well suited for intelligent RAN deployments. --- ### BERTO: an Adaptive BERT-based Network Time Series Predictor with Operator Preferences in Natural Language Authors: Nitin Priyadarshini Shankar, Vaibhav Singh, Sheetal Kalyani, Christian Maciocco Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05721v1 We introduce BERTO, a BERT-based framework for traffic prediction and energy optimization in cellular networks. Built on transformer architectures, BERTO delivers high prediction accuracy, while its Balancing Loss Function and prompt-based customization allow operators to adjust the trade-off between power savings and performance. Natural language prompts guide the model to manage underprediction and overprediction in accordance with the operator's intent. Experiments on real-world datasets show that BERTO improves upon existing models with a $4.13$\% reduction in MSE while introducing the feature of balancing competing objectives of power saving and performance through simple natural language inputs, operating over a flexible range of $1.4$ kW in power and up to $9\times$ variation in service quality, making it well suited for intelligent RAN deployments. --- ### Comparing the latent features of universal machine-learning interatomic potentials Authors: Sofiia Chorna, Davide Tisi, Cesare Malosso, Wei Bin How, Michele Ceriotti, Sanggyu Chong Published: 2025-12-05 Categories: physics.chem-ph, cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2512.05717v3 The past few years have seen the development of ``universal'' machine-learning interatomic potentials (uMLIPs) capable of approximating the ground-state potential energy surface across a wide range of chemical structures and compositions with reasonable accuracy. While these models differ in the architecture and the dataset used, they share the ability to compress a staggering amount of chemical information into descriptive latent features. Herein, we systematically analyze what the different uMLIPs have learned by quantitatively assessing the relative information content of their latent features with feature reconstruction errors, and observing how the trends are affected by the choice of training set and training protocol. We find that uMLIPs encode the chemical space in significantly distinct ways, with substantial cross-model feature reconstruction errors. When variants of the same model architecture are considered, trends become dependent on the dataset, target, and training protocol of choice. We also observe that fine-tuning of a uMLIP retains a strong pre-training bias in the latent features. Finally, we discuss how atom-level features, which are directly output by MLIPs, can be compressed into global structure-level features via concatenation of progressive cumulants, each adding significantly new information about the variability across the atomic environments within a given system. --- ### Comparing the latent features of universal machine-learning interatomic potentials Authors: Sofiia Chorna, Davide Tisi, Cesare Malosso, Wei Bin How, Michele Ceriotti, Sanggyu Chong Published: 2025-12-05 Categories: physics.chem-ph, cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2512.05717v1 The past few years have seen the development of ``universal'' machine-learning interatomic potentials (uMLIPs) capable of approximating the ground-state potential energy surface across a wide range of chemical structures and compositions with reasonable accuracy. While these models differ in the architecture and the dataset used, they share the ability to compress a staggering amount of chemical information into descriptive latent features. Herein, we systematically analyze what the different uMLIPs have learned by quantitatively assessing the relative information content of their latent features with feature reconstruction errors as metrics, and observing how the trends are affected by the choice of training set and training protocol. We find that the uMLIPs encode chemical space in significantly distinct ways, with substantial cross-model feature reconstruction errors. When variants of the same model architecture are considered, trends become dependent on the dataset, target, and training protocol of choice. We also observe that fine-tuning of a uMLIP retains a strong pre-training bias in the latent features. Finally, we discuss how atom-level features, which are directly output by MLIPs, can be compressed into global structure-level features via concatenation of progressive cumulants, each adding significantly new information about the variability across the atomic environments within a given system. --- ### Big Tech-Funded AI Papers Have Higher Citation Impact, Greater Insularity, and Larger Recency Bias Authors: Max Martin Gnewuch, Jan Philip Wahle, Terry Ruas, Bela Gipp Published: 2025-12-05 Categories: cs.DL, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2512.05714v1 Over the past four decades, artificial intelligence (AI) research has flourished at the nexus of academia and industry. However, Big Tech companies have increasingly acquired the edge in computational resources, big data, and talent. So far, it has been largely unclear how many papers the industry funds, how their citation impact compares to non-funded papers, and what drives industry interest. This study fills that gap by quantifying the number of industry-funded papers at 10 top AI conferences (e.g., ICLR, CVPR, AAAI, ACL) and their citation influence. We analyze about 49.8K papers, about 1.8M citations from AI papers to other papers, and about 2.3M citations from other papers to AI papers from 1998-2022 in Scopus. Through seven research questions, we examine the volume and evolution of industry funding in AI research, the citation impact of funded papers, the diversity and temporal range of their citations, and the subfields in which industry predominantly acts. Our findings reveal that industry presence has grown markedly since 2015, from less than 2 percent to more than 11 percent in 2020. Between 2018 and 2022, 12 percent of industry-funded papers achieved high citation rates as measured by the h5-index, compared to 4 percent of non-industry-funded papers and 2 percent of non-funded papers. Top AI conferences engage more with industry-funded research than non-funded research, as measured by our newly proposed metric, the Citation Preference Ratio (CPR). We show that industry-funded research is increasingly insular, citing predominantly other industry-funded papers while referencing fewer non-funded papers. These findings reveal new trends in AI research funding, including a shift towards more industry-funded papers and their growing citation impact, greater insularity of industry-funded work than non-funded work, and a preference of industry-funded research to cite recent work. --- ### Bayesian Active Inference for Intelligent UAV Anti-Jamming and Adaptive Trajectory Planning Authors: Ali Krayani, Seyedeh Fatemeh Sadati, Lucio Marcenaro, Carlo Regazzoni Published: 2025-12-05 Categories: cs.RO, cs.AI, eess.SP, eess.SY arXiv: https://arxiv.org/abs/2512.05711v1 This paper proposes a hierarchical trajectory planning framework for UAVs operating under adversarial jamming conditions. Leveraging Bayesian Active Inference, the approach combines expert-generated demonstrations with probabilistic generative modeling to encode high-level symbolic planning, low-level motion policies, and wireless signal feedback. During deployment, the UAV performs online inference to anticipate interference, localize jammers, and adapt its trajectory accordingly, without prior knowledge of jammer locations. Simulation results demonstrate that the proposed method achieves near-expert performance, significantly reducing communication interference and mission cost compared to model-free reinforcement learning baselines, while maintaining robust generalization in dynamic environments. --- ### Manifold-Aware Point Cloud Completion via Geodesic-Attentive Hierarchical Feature Learning Authors: Jianan Sun, Dongzhihan Wang, Mingyu Fan Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05710v1 Point cloud completion seeks to recover geometrically consistent shapes from partial or sparse 3D observations. Although recent methods have achieved reasonable global shape reconstruction, they often rely on Euclidean proximity and overlook the intrinsic nonlinear geometric structure of point clouds, resulting in suboptimal geometric consistency and semantic ambiguity. In this paper, we present a manifold-aware point cloud completion framework that explicitly incorporates nonlinear geometry information throughout the feature learning pipeline. Our approach introduces two key modules: a Geodesic Distance Approximator (GDA), which estimates geodesic distances between points to capture the latent manifold topology, and a Manifold-Aware Feature Extractor (MAFE), which utilizes geodesic-based $k$-NN groupings and a geodesic-relational attention mechanism to guide the hierarchical feature extraction process. By integrating geodesic-aware relational attention, our method promotes semantic coherence and structural fidelity in the reconstructed point clouds. Extensive experiments on benchmark datasets demonstrate that our approach consistently outperforms state-of-the-art methods in reconstruction quality. --- ### Faithfulness metric fusion: Improving the evaluation of LLM trustworthiness across domains Authors: Ben Malin, Tatiana Kalganova, Nikolaos Boulgouris Published: 2025-12-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.05700v1 We present a methodology for improving the accuracy of faithfulness evaluation in Large Language Models (LLMs). The proposed methodology is based on the combination of elementary faithfulness metrics into a combined (fused) metric, for the purpose of improving the faithfulness of LLM outputs. The proposed strategy for metric fusion deploys a tree-based model to identify the importance of each metric, which is driven by the integration of human judgements evaluating the faithfulness of LLM responses. This fused metric is demonstrated to correlate more strongly with human judgements across all tested domains for faithfulness. Improving the ability to evaluate the faithfulness of LLMs, allows for greater confidence to be placed within models, allowing for their implementation in a greater diversity of scenarios. Additionally, we homogenise a collection of datasets across question answering and dialogue-based domains and implement human judgements and LLM responses within this dataset, allowing for the reproduction and trialling of faithfulness evaluation across domains. --- ### OWL: Unsupervised 3D Object Detection by Occupancy Guided Warm-up and Large Model Priors Reasoning Authors: Xusheng Guo, Wanfa Zhang, Shijia Zhao, Qiming Xia, Xiaolong Xie, Mingming Wang, Hai Wu, Chenglu Wen Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05698v1 Unsupervised 3D object detection leverages heuristic algorithms to discover potential objects, offering a promising route to reduce annotation costs in autonomous driving. Existing approaches mainly generate pseudo labels and refine them through self-training iterations. However, these pseudo-labels are often incorrect at the beginning of training, resulting in misleading the optimization process. Moreover, effectively filtering and refining them remains a critical challenge. In this paper, we propose OWL for unsupervised 3D object detection by occupancy guided warm-up and large-model priors reasoning. OWL first employs an Occupancy Guided Warm-up (OGW) strategy to initialize the backbone weight with spatial perception capabilities, mitigating the interference of incorrect pseudo-labels on network convergence. Furthermore, OWL introduces an Instance-Cued Reasoning (ICR) module that leverages the prior knowledge of large models to assess pseudo-label quality, enabling precise filtering and refinement. Finally, we design a Weight-adapted Self-training (WAS) strategy to dynamically re-weight pseudo-labels, improving the performance through self-training. Extensive experiments on Waymo Open Dataset (WOD) and KITTI demonstrate that OWL outperforms state-of-the-art unsupervised methods by over 15.0% mAP, revealing the effectiveness of our method. --- ### HiMoE-VLA: Hierarchical Mixture-of-Experts for Generalist Vision-Language-Action Policies Authors: Zhiying Du, Bei Liu, Yaobo Liang, Yichao Shen, Haidong Cao, Xiangyu Zheng, Zhiyuan Feng, Zuxuan Wu, Jiaolong Yang, Yu-Gang Jiang Published: 2025-12-05 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2512.05693v2 Generalist vision--language--action (VLA) policies are typically trained on heterogeneous mixtures of robot demonstrations spanning diverse embodiments, action spaces, and observation configurations. Modeling such heterogeneity with a shared dense action module can induce negative transfer, particularly when action spaces or visual observations differ across data sources. We address this issue with HiMoE-VLA, a VLA framework built around a Hierarchical Mixture-of-Experts (HiMoE) action module. HiMoE uses Action-Space MoE layers at the input/output boundaries to specialize computation for distinct action spaces, Heterogeneity-Balancing MoE layers in neighboring layers to provide balanced capacity for residual variation in observations, scenes, and embodiments, and dense Transformer blocks in the middle to integrate shared representations. Two auxiliary objectives further guide this hierarchy: a contrastive Action-Space Regularization objective for boundary specialization and a load-balancing objective for stable expert utilization. HiMoE-VLA reaches 3.98 on CALVIN, 98.0\% on LIBERO, and 75.0\% and 63.7\% average success on real xArm7 and ALOHA tasks; under controlled heterogeneous co-training, it turns the negative transfer observed in strong baselines into positive transfer. The code and models are publicly available at https://github.com/ZhiyingDu/HiMoE-VLA. --- ### HiMoE-VLA: Hierarchical Mixture-of-Experts for Generalist Vision-Language-Action Policies Authors: Zhiying Du, Bei Liu, Yaobo Liang, Yichao Shen, Haidong Cao, Xiangyu Zheng, Zhiyuan Feng, Zuxuan Wu, Jiaolong Yang, Yu-Gang Jiang Published: 2025-12-05 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2512.05693v1 The development of foundation models for embodied intelligence critically depends on access to large-scale, high-quality robot demonstration data. Recent approaches have sought to address this challenge by training on large collections of heterogeneous robotic datasets. However, unlike vision or language data, robotic demonstrations exhibit substantial heterogeneity across embodiments and action spaces as well as other prominent variations such as senor configurations and action control frequencies. The lack of explicit designs for handling such heterogeneity causes existing methods to struggle with integrating diverse factors, thereby limiting their generalization and leading to degraded performance when transferred to new settings. In this paper, we present HiMoE-VLA, a novel vision-language-action (VLA) framework tailored to effectively handle diverse robotic data with heterogeneity. Specifically, we introduce a Hierarchical Mixture-of-Experts (HiMoE) architecture for the action module which adaptively handles multiple sources of heterogeneity across layers and gradually abstracts them into shared knowledge representations. Through extensive experimentation with simulation benchmarks and real-world robotic platforms, HiMoE-VLA demonstrates a consistent performance boost over existing VLA baselines, achieving higher accuracy and robust generalization across diverse robots and action spaces. The code and models are publicly available at https://github.com/ZhiyingDu/HiMoE-VLA. --- ### Physics-Informed Graph Neural Networks for Frequency-Aware Optical Aberration Correction Authors: Yong En Kok, Bowen Deng, Alexander Bentley, Andrew J. Parkes, Michael G. Somekh, Amanda J. Wright, Michael P. Pound Published: 2025-12-05 Categories: cs.CV, physics.optics arXiv: https://arxiv.org/abs/2512.05683v2 Optical aberrations significantly degrade image quality in microscopy, particularly when imaging deeper into samples. These aberrations arise from distortions in the optical wavefront and can be mathematically represented using Zernike polynomials. Existing methods often address only mild aberrations on limited sample types and modalities, typically treating the problem as a black-box mapping without leveraging the underlying optical physics of wavefront distortions. We propose ZRNet, a physics-informed framework that jointly performs Zernike coefficient prediction and optical image Restoration. We contribute a Zernike Graph module that explicitly models physical relationships between Zernike polynomials based on their azimuthal degrees-ensuring that learned corrections align with fundamental optical principles. To further enforce physical consistency between image restoration and Zernike prediction, we introduce a Frequency-Aware Alignment (FAA) loss, which better aligns Zernike coefficient prediction and image features in the Fourier domain. Extensive experiments on CytoImageNet demonstrates that our approach achieves state-of-the-art performance in both image restoration and Zernike coefficient prediction across diverse microscopy modalities and biological samples with complex, large-amplitude aberrations. We further validate on experimental PSF data from a physical microscope and demonstrate robustness to realistic sensor noise, confirming generalisation beyond simulated conditions. Code is available at https://github.com/janetkok/ZRNet. --- ### Physics-Informed Graph Neural Network with Frequency-Aware Learning for Optical Aberration Correction Authors: Yong En Kok, Bowen Deng, Alexander Bentley, Andrew J. Parkes, Michael G. Somekh, Amanda J. Wright, Michael P. Pound Published: 2025-12-05 Categories: cs.CV, physics.optics arXiv: https://arxiv.org/abs/2512.05683v1 Optical aberrations significantly degrade image quality in microscopy, particularly when imaging deeper into samples. These aberrations arise from distortions in the optical wavefront and can be mathematically represented using Zernike polynomials. Existing methods often address only mild aberrations on limited sample types and modalities, typically treating the problem as a black-box mapping without leveraging the underlying optical physics of wavefront distortions. We propose ZRNet, a physics-informed framework that jointly performs Zernike coefficient prediction and optical image Restoration. We contribute a Zernike Graph module that explicitly models physical relationships between Zernike polynomials based on their azimuthal degrees-ensuring that learned corrections align with fundamental optical principles. To further enforce physical consistency between image restoration and Zernike prediction, we introduce a Frequency-Aware Alignment (FAA) loss, which better aligns Zernike coefficient prediction and image features in the Fourier domain. Extensive experiments on CytoImageNet demonstrates that our approach achieves state-of-the-art performance in both image restoration and Zernike coefficient prediction across diverse microscopy modalities and biological samples with complex, large-amplitude aberrations. Code is available at https://github.com/janetkok/ZRNet. --- ### Retrieving Semantically Similar Decisions under Noisy Institutional Labels: Robust Comparison of Embedding Methods Authors: Tereza Novotna, Jakub Harasta Published: 2025-12-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.05681v1 Retrieving case law is a time-consuming task predominantly carried out by querying databases. We provide a comparison of two models in three different settings for Czech Constitutional Court decisions: (i) a large general-purpose embedder (OpenAI), (ii) a domain-specific BERT-trained from scratch on ~30,000 decisions using sliding windows and attention pooling. We propose a noise-aware evaluation including IDF-weighted keyword overlap as graded relevance, binarization via two thresholds (0.20 balanced, 0.28 strict), significance via paired bootstrap, and an nDCG diagnosis supported with qualitative analysis. Despite modest absolute nDCG (expected under noisy labels), the general OpenAI embedder decisively outperforms the domain pre-trained BERT in both settings at @10/@20/@100 across both thresholds; differences are statistically significant. Diagnostics attribute low absolutes to label drift and strong ideals rather than lack of utility. Additionally, our framework is robust enough to be used for evaluation under a noisy gold dataset, which is typical when handling data with heterogeneous labels stemming from legacy judicial databases. --- ### Meta-Learning Multi-armed Bandits for Beam Tracking in 5G and 6G Networks Authors: Alexander Mattick, George Yammine, Georgios Kontes, Setareh Maghsudi, Christopher Mutschler Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05680v1 Beamforming-capable antenna arrays with many elements enable higher data rates in next generation 5G and 6G networks. In current practice, analog beamforming uses a codebook of pre-configured beams with each of them radiating towards a specific direction, and a beam management function continuously selects \textit{optimal} beams for moving user equipments (UEs). However, large codebooks and effects caused by reflections or blockages of beams make an optimal beam selection challenging. In contrast to previous work and standardization efforts that opt for supervised learning to train classifiers to predict the next best beam based on previously selected beams we formulate the problem as a partially observable Markov decision process (POMDP) and model the environment as the codebook itself. At each time step, we select a candidate beam conditioned on the belief state of the unobservable optimal beam and previously probed beams. This frames the beam selection problem as an online search procedure that locates the moving optimal beam. In contrast to previous work, our method handles new or unforeseen trajectories and changes in the physical environment, and outperforms previous work by orders of magnitude. --- ### Empirical Decision Theory Authors: Christoph Jansen, Georg Schollmeyer, Thomas Augustin, Julian Rodemann Published: 2025-12-05 Categories: stat.ME, math.PR, stat.ML arXiv: https://arxiv.org/abs/2512.05677v1 Analyzing decision problems under uncertainty commonly relies on idealizing assumptions about the describability of the world, with the most prominent examples being the closed world and the small world assumption. Most assumptions are operationalized by introducing states of the world, conditional on which the decision situation can be analyzed without any remaining uncertainty. Conversely, most classical decision-theoretic approaches are not applicable if the states of the world are inaccessible. We propose a decision model that retains the appeal and simplicity of the original theory, but completely overcomes the need to specify the states of the world explicitly. The main idea of our approach is to address decision problems in a radically empirical way: instead of specifying states and consequences prior to the decision analysis, we only assume a protocol of observed act--consequence pairs as model primitives. We show how optimality in such empirical decision problems can be addressed by using protocol-based empirical choice functions and discuss three approaches for deriving inferential guarantees: (I) consistent statistical estimation of choice sets, (II) consistent statistical testing of choice functions with robustness guarantees, and (III) direct inference for empirical choice functions using credal sets. We illustrate our theory with a proof-of-concept application comparing different prompting strategies in generative AI models. --- ### Hyperspectral Unmixing with 3D Convolutional Sparse Coding and Projected Simplex Volume Maximization Authors: Gargi Panda, Soumitra Kundu, Saumik Bhattacharya, Aurobinda Routray Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05674v1 Hyperspectral unmixing (HSU) aims to separate each pixel into its constituent endmembers and estimate their corresponding abundance fractions. This work presents an algorithm-unrolling-based network for the HSU task, named the 3D Convolutional Sparse Coding Network (3D-CSCNet), built upon a 3D CSC model. Unlike existing unrolling-based networks, our 3D-CSCNet is designed within the powerful autoencoder (AE) framework. Specifically, to solve the 3D CSC problem, we propose a 3D CSC block (3D-CSCB) derived through deep algorithm unrolling. Given a hyperspectral image (HSI), 3D-CSCNet employs the 3D-CSCB to estimate the abundance matrix. The use of 3D CSC enables joint learning of spectral and spatial relationships in the 3D HSI data cube. The estimated abundance matrix is then passed to the AE decoder to reconstruct the HSI, and the decoder weights are extracted as the endmember matrix. Additionally, we propose a projected simplex volume maximization (PSVM) algorithm for endmember estimation, and the resulting endmembers are used to initialize the decoder weights of 3D-CSCNet. Extensive experiments on three real datasets and one simulated dataset with three different signal-to-noise ratio (SNR) levels demonstrate that our 3D-CSCNet outperforms state-of-the-art methods. --- ### InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem Authors: Yeobin Hong, Suhyeon Lee, Hyungjin Chung, Jong Chul Ye Published: 2025-12-05 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.05672v2 Recent approaches in controllable novel view video generation often rely on fine-tuning pre-trained Video Diffusion Models (VDMs). This dominant paradigm is computationally expensive and frequently suffers from catastrophic forgetting of the model's original generative priors. To address this challenge, here we propose InverseCrafter, a VDM training-free framework that reformulates novel view video generation as an inpainting-based inverse problem in the latent space, eliminating the need for any annotated 4D training data. The core of our method is to establish operator equivalence by employing a lightweight latent mask encoder to define a latent-domain masking operation via a continuous, multi-channel representation. This principled representation faithfully models the forward process in the latent domain, enabling efficient, backpropagation-free solvers while bypassing the costly bottleneck of repeated VAE operations. InverseCrafter achieves high-fidelity, spatio-temporally coherent novel view synthesis with near-zero additional inference overhead and excels at general-purpose video inpainting and editing by fully preserving the pre-trained VDM's generative capabilities. --- ### InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem Authors: Yeobin Hong, Suhyeon Lee, Hyungjin Chung, Jong Chul Ye Published: 2025-12-05 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.05672v1 Recent approaches to controllable 4D video generation often rely on fine-tuning pre-trained Video Diffusion Models (VDMs). This dominant paradigm is computationally expensive, requiring large-scale datasets and architectural modifications, and frequently suffers from catastrophic forgetting of the model's original generative priors. Here, we propose InverseCrafter, an efficient inpainting inverse solver that reformulates the 4D generation task as an inpainting problem solved in the latent space. The core of our method is a principled mechanism to encode the pixel space degradation operator into a continuous, multi-channel latent mask, thereby bypassing the costly bottleneck of repeated VAE operations and backpropagation. InverseCrafter not only achieves comparable novel view generation and superior measurement consistency in camera control tasks with near-zero computational overhead, but also excels at general-purpose video inpainting with editing. Code is available at https://github.com/yeobinhong/InverseCrafter. --- ### ClinTutor-R1: Advancing Scalable and Robust One-to-Many Alignment in Clinical Socratic Education Authors: Zhitao He, Haolin Yang, Zeyu Qin, Yi R Fung Published: 2025-12-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.05671v2 While Large Language Models (LLMs) have achieved remarkable success in dyadic (one-on-one) instruction, they face significant challenges in One-to-Many alignment, such as clinical ward rounds, where an instructor must simultaneously guide a diverse group of trainees. Current models often suffer from context dilution and goal misalignment, failing to balance individual scaffolding with collective learning progress. To address this, we introduce ClinEdu, a multi-agent pedagogical simulator that models the complexity of group dynamics. Leveraging this platform, we construct ClinTeach, a large-scale dataset of Socratic teaching dialogues, and propose ClinTutor-R1, the first vision-language agent explicitly architected to achieve one-to-many alignment in clinical education, employing an explicit internal thinking mechanism to model both individual belief states and group consensus. We validate our framework through a comprehensive protocol covering static benchmarks, in-situ interactive evaluation within ClinEdu, expert assessment, and a 200-participant real user study. Experimental results demonstrate that ClinTutor-R1 outperforms base models by over 20% and achieves parity with proprietary models, while exhibiting scalability in maintaining instructional quality across expanding student cohorts. --- ### MedTutor-R1: Socratic Personalized Medical Teaching with Multi-Agent Simulation Authors: Zhitao He, Haolin Yang, Zeyu Qin, Yi R Fung Published: 2025-12-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.05671v1 The significant gap between rising demands for clinical training and the scarcity of expert instruction poses a major challenge to medical education. With powerful capabilities in personalized guidance, Large Language Models (LLMs) offer a promising solution to bridge this gap. However, current research focuses mainly on one-on-one knowledge instruction, overlooking collaborative reasoning, a key skill for students developed in teamwork like ward rounds. To this end, we develop ClinEdu, a multi-agent pedagogical simulator with personality-driven patients and diverse student cohorts, enabling controlled testing of complex pedagogical processes and scalable generation of teaching data. Based on ClinEdu, we construct ClinTeach, a large Socratic teaching dialogue dataset that captures the complexities of group instruction. We then train MedTutor-R1, the first multimodal Socratic tutor designed for one-to-many instruction in clinical medical education. MedTutor-R1 is first instruction-tuned on our ClinTeach dataset and then optimized with reinforcement learning, using rewards derived from a three-axis rubric, covering structural fidelity, analytical quality, and clinical safety, to refine its adaptive Socratic strategies. For authentic in-situ assessment, we use simulation-based interactive evaluation that redeploys the tutor back into ClinEdu. Experimental results demonstrate that our MedTutor-R1 outperforms the base model by over 20% in average pedagogical score and is comparable to o3, while also exhibiting high adaptability in handling a varying number of students. This promising performance underscores the effectiveness of our pedagogical simulator, ClinEdu. --- ### Deep Learning-Based Real-Time Sequential Facial Expression Analysis Using Geometric Features Authors: Talha Enes Koksal, Abdurrahman Gumus Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05669v1 Facial expression recognition is a crucial component in enhancing human-computer interaction and developing emotion-aware systems. Real-time detection and interpretation of facial expressions have become increasingly important for various applications, from user experience personalization to intelligent surveillance systems. This study presents a novel approach to real-time sequential facial expression recognition using deep learning and geometric features. The proposed method utilizes MediaPipe FaceMesh for rapid and accurate facial landmark detection. Geometric features, including Euclidean distances and angles, are extracted from these landmarks. Temporal dynamics are incorporated by analyzing feature differences between consecutive frames, enabling the detection of onset, apex, and offset phases of expressions. For classification, a ConvLSTM1D network followed by multilayer perceptron blocks is employed. The method's performance was evaluated on multiple publicly available datasets, including CK+, Oulu-CASIA (VIS and NIR), and MMI. Accuracies of 93%, 79%, 77%, and 68% were achieved respectively. Experiments with composite datasets were also conducted to assess the model's generalization capabilities. The approach demonstrated real-time applicability, processing approximately 165 frames per second on consumer-grade hardware. This research contributes to the field of facial expression analysis by providing a fast, accurate, and adaptable solution. The findings highlight the potential for further advancements in emotion-aware technologies and personalized user experiences, paving the way for more sophisticated human-computer interaction systems. To facilitate further research in this field, the complete source code for this study has been made publicly available on GitHub: https://github.com/miralab-ai/facial-expression-analysis. --- ### On Dynamic Programming Theory for Leader-Follower Stochastic Games Authors: Jilles Steeve Dibangoye, Thibaut Le Marre, Ocan Sankur, François Schwarzentruber Published: 2025-12-05 Categories: cs.GT, cs.AI arXiv: https://arxiv.org/abs/2512.05667v1 Leader-follower general-sum stochastic games (LF-GSSGs) model sequential decision-making under asymmetric commitment, where a leader commits to a policy and a follower best responds, yielding a strong Stackelberg equilibrium (SSE) with leader-favourable tie-breaking. This paper introduces a dynamic programming (DP) framework that applies Bellman recursion over credible sets-state abstractions formally representing all rational follower best responses under partial leader commitments-to compute SSEs. We first prove that any LF-GSSG admits a lossless reduction to a Markov decision process (MDP) over credible sets. We further establish that synthesising an optimal memoryless deterministic leader policy is NP-hard, motivating the development of ε-optimal DP algorithms with provable guarantees on leader exploitability. Experiments on standard mixed-motive benchmarks-including security games, resource allocation, and adversarial planning-demonstrate empirical gains in leader value and runtime scalability over state-of-the-art methods. --- ### Feasibility of AI-Assisted Programming for End-User Development Authors: Irene Weber Published: 2025-12-05 Categories: cs.LG, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2512.05666v1 End-user development,where non-programmers create or adapt their own digital tools, can play a key role in driving digital transformation within organizations. Currently, low-code/no-code platforms are widely used to enable end-user development through visual programming, minimizing the need for manual coding. Recent advancements in generative AI, particularly large language model-based assistants and "copilots", open new possibilities, as they may enable end users to generate and refine programming code and build apps directly from natural language prompts. This approach, here referred to as AI-assisted end-user coding, promises greater flexibility, broader applicability, faster development, improved reusability, and reduced vendor lock-in compared to the established visual LCNC platforms. This paper investigates whether AI-assisted end-user coding is a feasible paradigm for end-user development, which may complement or even replace the LCNC model in the future. To explore this, we conducted a case study in which non-programmers were asked to develop a basic web app through interaction with AI assistants.The majority of study participants successfully completed the task in reasonable time and also expressed support for AI-assisted end-user coding as a viable approach for end-user development. The paper presents the study design, analyzes the outcomes, and discusses potential implications for practice, future research, and academic teaching. --- ### Interleaved Latent Visual Reasoning with Selective Perceptual Modeling Authors: Shuai Dong, Siyuan Wang, Xingyu Liu, Chenglin Li, Haowen Hou, Zhongyu Wei Published: 2025-12-05 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2512.05665v3 Interleaved reasoning paradigms enhance Multimodal Large Language Models (MLLMs) with visual feedback but are hindered by the prohibitive computational cost of re-encoding pixel-dense images. A promising alternative, latent visual reasoning, circumvents this bottleneck yet faces limitations: methods either fail to capture intermediate state evolution due to single-step, non-interleaved structures, or sacrifice precise perceptual modeling by over-compressing features. We introduce Interleaved Latent Visual Reasoning (ILVR), a framework that unifies dynamic state evolution with precise perceptual modeling. ILVR interleaves textual generation with latent visual representations that act as specific, evolving cues for subsequent reasoning. Specifically, we employ a self-supervision strategy where a momentum teacher model selectively distills relevant features from ground-truth intermediate images into sparse supervision targets. This adaptive selection mechanism guides the model to autonomously generate context-aware visual signals. Extensive experiments on multimodal reasoning benchmarks demonstrate that ILVR outperforms existing approaches, effectively bridging the gap between fine-grained perception and sequential multimodal reasoning. The code is available at https://github.com/XD111ds/ILVR. --- ### Interleaved Latent Visual Reasoning with Selective Perceptual Modeling Authors: Shuai Dong, Siyuan Wang, Xingyu Liu, Zhongyu Wei Published: 2025-12-05 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2512.05665v1 Interleaved reasoning paradigms enhance Multimodal Large Language Models (MLLMs) with visual feedback but are hindered by the prohibitive computational cost of repeatedly re-encoding pixel-dense images. A promising alternative, latent visual reasoning, circumvents this bottleneck yet currently forces a critical trade-off: methods either sacrifice precise perceptual modeling by over-compressing features or fail to model dynamic problems due to static, non-interleaved structures. We introduce Interleaved Latent Visual Reasoning (ILVR), a framework that unifies dynamic state evolution with precise perceptual modeling. ILVR interleaves textual generation with latent visual representations that act as specific, evolving cues for subsequent reasoning. To enable this, we employ a self-supervision strategy where a Momentum Teacher Model selectively distills relevant features from helper images into sparse supervision targets. This adaptive selection mechanism guides the model to autonomously generate context-aware visual signals. Extensive experiments on multimodal reasoning benchmarks demonstrate that ILVR significantly outperforms existing approaches, effectively bridging the gap between fine-grained perception and sequential multimodal reasoning. --- ### LeAD-M3D: Leveraging Asymmetric Distillation for Real-Time Monocular 3D Detection Authors: Johannes Meier, Jonathan Michel, Oussema Dhaouadi, Yung-Hsu Yang, Christoph Reich, Zuria Bauer, Stefan Roth, Marc Pollefeys, Jacques Kaiser, Daniel Cremers Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05663v3 Real-time monocular 3D object detection remains challenging due to severe depth ambiguity, viewpoint shifts, and the high computational cost of 3D reasoning. Existing approaches either rely on LiDAR or geometric priors to compensate for missing depth or sacrifice efficiency to achieve competitive accuracy. We introduce LeAD-M3D, a monocular 3D detector that achieves state-of-the-art accuracy and real-time inference without extra modalities. Our method is enabled by three key components. Asymmetric Augmentation Denoising Distillation (A2D2) transfers geometric knowledge from a clean-image teacher to a MixUp-noised student via a quality- and importance-weighted depth-feature loss, enabling stronger depth reasoning without LiDAR. 3D-aware Consistent Matching (CM$_{\text{3D}}$) improves prediction-to-ground truth assignment by integrating 3D MGIoU into the matching score, yielding stable and precise supervision. Finally, Confidence-Gated 3D Inference (CGI$_{\text{3D}}$) accelerates inference by restricting expensive 3D regression to confident regions. Together, these contributions set a new Pareto frontier for monocular 3D detection: LeAD-M3D achieves state-of-the-art accuracy on KITTI and Waymo, and the best reported car AP on Rope3D, while running up to 3.6$\,\times$ faster than prior high-accuracy models (e.g., MonoDiff). LeAD-M3D demonstrates that high fidelity and real-time monocular 3D detection is simultaneously attainable, without LiDAR, stereo, or strong geometric assumptions. --- ### LeAD-M3D: Leveraging Asymmetric Distillation for Real-time Monocular 3D Detection Authors: Johannes Meier, Jonathan Michel, Oussema Dhaouadi, Yung-Hsu Yang, Christoph Reich, Zuria Bauer, Stefan Roth, Marc Pollefeys, Jacques Kaiser, Daniel Cremers Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05663v1 Real-time monocular 3D object detection remains challenging due to severe depth ambiguity, viewpoint shifts, and the high computational cost of 3D reasoning. Existing approaches either rely on LiDAR or geometric priors to compensate for missing depth, or sacrifice efficiency to achieve competitive accuracy. We introduce LeAD-M3D, a monocular 3D detector that achieves state-of-the-art accuracy and real-time inference without extra modalities. Our method is powered by three key components. Asymmetric Augmentation Denoising Distillation (A2D2) transfers geometric knowledge from a clean-image teacher to a mixup-noised student via a quality- and importance-weighted depth-feature loss, enabling stronger depth reasoning without LiDAR supervision. 3D-aware Consistent Matching (CM3D) improves prediction-to-ground truth assignment by integrating 3D MGIoU into the matching score, yielding more stable and precise supervision. Finally, Confidence-Gated 3D Inference (CGI3D) accelerates detection by restricting expensive 3D regression to top-confidence regions. Together, these components set a new Pareto frontier for monocular 3D detection: LeAD-M3D achieves state-of-the-art accuracy on KITTI and Waymo, and the best reported car AP on Rope3D, while running up to 3.6x faster than prior high-accuracy methods. Our results demonstrate that high fidelity and real-time efficiency in monocular 3D detection are simultaneously attainable - without LiDAR, stereo, or geometric assumptions. --- ### Multilingual Medical Reasoning for Question Answering with Large Language Models Authors: Pietro Ferrazzi, Aitor Soroa, Rodrigo Agerri Published: 2025-12-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.05658v3 Large Language Models (LLMs) with reasoning capabilities have recently demonstrated strong potential in medical Question Answering (QA). Existing approaches are largely English-focused and primarily rely on distillation from general-purpose LLMs, raising concerns about the reliability of their medical knowledge. In this work, we present a method to generate multilingual reasoning traces based on medical knowledge extracted from Wikipedia. We produce 500k traces in English, Italian, and Spanish, using a retrieval-augmented generation approach over medical information from Wikipedia. The traces are generated to solve medical questions drawn from MedQA and MedMCQA, which we extend to Italian and Spanish. We test our pipeline in both in-domain and out-of-domain settings across Medical QA benchmarks, and demonstrate that our reasoning traces improve performance both when utilized via in-context learning (few-shot) and supervised fine-tuning, yielding state-of-the-art results among 8B-parameter LLMs. We believe that these resources can support the development of more transparent clinical decision-support tools in multilingual settings. We release the full suite of resources: reasoning traces, translated QA datasets, Medical-Wikipedia, and fine-tuned models. --- ### Grounded Multilingual Medical Reasoning for Question Answering with Large Language Models Authors: Pietro Ferrazzi, Aitor Soroa, Rodrigo Agerri Published: 2025-12-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.05658v1 Large Language Models (LLMs) with reasoning capabilities have recently demonstrated strong potential in medical Question Answering (QA). Existing approaches are largely English-focused and primarily rely on distillation from general-purpose LLMs, raising concerns about the reliability of their medical knowledge. In this work, we present a method to generate multilingual reasoning traces grounded in factual medical knowledge. We produce 500k traces in English, Italian, and Spanish, using a retrievalaugmented generation approach over medical information from Wikipedia. The traces are generated to solve medical questions drawn from MedQA and MedMCQA, which we extend to Italian and Spanish. We test our pipeline in both in-domain and outof-domain settings across Medical QA benchmarks, and demonstrate that our reasoning traces improve performance both when utilized via in-context learning (few-shot) and supervised fine-tuning, yielding state-of-the-art results among 8B-parameter LLMs. We believe that these resources can support the development of safer, more transparent clinical decision-support tools in multilingual settings. We release the full suite of resources: reasoning traces, translated QA datasets, Medical-Wikipedia, and fine-tuned models. --- ### Over-the-Air Semantic Alignment with Stacked Intelligent Metasurfaces Authors: Mario Edoardo Pandolfo, Kyriakos Stylianopoulos, George C. Alexandropoulos, Paolo Di Lorenzo Published: 2025-12-05 Categories: eess.SP, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2512.05657v1 Semantic communication systems aim to transmit task-relevant information between devices capable of artificial intelligence, but their performance can degrade when heterogeneous transmitter-receiver models produce misaligned latent representations. Existing semantic alignment methods typically rely on additional digital processing at the transmitter or receiver, increasing overall device complexity. In this work, we introduce the first over-the-air semantic alignment framework based on stacked intelligent metasurfaces (SIM), which enables latent-space alignment directly in the wave domain, reducing substantially the computational burden at the device level. We model SIMs as trainable linear operators capable of emulating both supervised linear aligners and zero-shot Parseval-frame-based equalizers. To realize these operators physically, we develop a gradient-based optimization procedure that tailors the metasurface transfer function to a desired semantic mapping. Experiments with heterogeneous vision transformer (ViT) encoders show that SIMs can accurately reproduce both supervised and zero-shot semantic equalizers, achieving up to 90% task accuracy in regimes with high signal-to-noise ratio (SNR), while maintaining strong robustness even at low SNR values. --- ### Self-Supervised AI-Generated Image Detection: A Camera Metadata Perspective Authors: Nan Zhong, Mian Zou, Yiran Xu, Zhenxing Qian, Xinpeng Zhang, Baoyuan Wu, Kede Ma Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05651v2 The proliferation of AI-generated imagery poses escalating challenges for multimedia forensics, yet many existing detectors depend on assumptions about the internals of specific generative models, limiting their cross-model applicability. We introduce a self-supervised approach for detecting AI-generated images that leverages camera metadata -- specifically exchangeable image file format (EXIF) tags -- to learn features intrinsic to digital photography. Our pretext task trains a feature extractor solely on camera-captured photographs by classifying categorical EXIF tags (\emph{e.g.}, camera model and scene type) and pairwise-ranking ordinal and continuous EXIF tags (\emph{e.g.}, focal length and aperture value). Using these EXIF-induced features, we first perform one-class detection by modeling the distribution of photographic images with a Gaussian mixture model and flagging low-likelihood samples as AI-generated. We then extend to binary detection that treats the learned extractor as a strong regularizer for a classifier of the same architecture, operating on high-frequency residuals from spatially scrambled patches. Extensive experiments across various generative models demonstrate that our EXIF-induced detectors substantially advance the state of the art, delivering strong generalization to in-the-wild samples and robustness to common benign image perturbations. The code and model are publicly available at https://github.com/Ekko-zn/SDAIE. --- ### Beyond Data Filtering: Knowledge Localization for Capability Removal in LLMs Authors: Igor Shilov, Alex Cloud, Aryo Pradipta Gema, Jacob Goldman-Wetzler, Nina Panickssery, Henry Sleight, Erik Jones, Cem Anil Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05648v1 Large Language Models increasingly possess capabilities that carry dual-use risks. While data filtering has emerged as a pretraining-time mitigation, it faces significant challenges: labeling whether data is harmful is expensive at scale, and given improving sample efficiency with larger models, even small amounts of mislabeled content could give rise to dangerous capabilities. To address risks associated with mislabeled harmful content, prior work proposed Gradient Routing (Cloud et al., 2024) -- a technique that localizes target knowledge into a dedicated subset of model parameters so they can later be removed. We explore an improved variant of Gradient Routing, which we call Selective GradienT Masking (SGTM), with particular focus on evaluating its robustness to label noise. SGTM zero-masks selected gradients such that target domain examples only update their dedicated parameters. We test SGTM's effectiveness in two applications: removing knowledge of one language from a model trained on a bilingual synthetic dataset, and removing biology knowledge from a model trained on English Wikipedia. In both cases SGTM provides better retain/forget trade-off in the presence of labeling errors compared to both data filtering and a previously proposed instantiation of Gradient Routing. Unlike shallow unlearning approaches that can be quickly undone through fine-tuning, SGTM exhibits strong robustness to adversarial fine-tuning, requiring seven times more fine-tuning steps to reach baseline performance on the forget set compared to a finetuning-based unlearning method (RMU). Our results suggest SGTM provides a promising pretraining-time complement to existing safety mitigations, particularly in settings where label noise is unavoidable. --- ### A Greek Government Decisions Dataset for Public-Sector Analysis and Insight Authors: Giorgos Antoniou, Giorgos Filandrianos, Aggelos Vlachos, Giorgos Stamou, Lampros Kollimenos, Konstantinos Skianis, Michalis Vazirgiannis Published: 2025-12-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.05647v2 We introduce an open, machine-readable corpus of Greek government decisions sourced from the national transparency platform Diavgeia. The resource comprises 1 million decisions, featuring and high-quality raw text extracted from PDFs. It is released with raw extracted text in Markdown format, alongside a fully reproducible extraction pipeline. Beyond the core dataset, we conduct qualitative analyses to explore boilerplate patterns and design a retrieval-augmented generation (RAG) task by formulating a set of representative questions, creating high-quality answers, and evaluating a baseline RAG system on its ability to retrieve and reason over public decisions. This evaluation demonstrates the potential of large-scale public-sector corpora to support advanced information access and transparency through structured retrieval and reasoning over governmental documents, and highlights how such a RAG pipeline could simulate a chat-based assistant capable of interactively answering questions about public decisions. Due to its scale, quality, and domain coverage, the corpus can also serve as high-value pre-training or fine-tuning material for new Language Models (LMs) and Large Language Models (LLMs) respectively, including specialized models for legal and governmental domains, and as a foundation for novel approaches in domain adaptation, knowledge-grounded generation, and explainable AI. Finally, we discuss limitations, outline future directions, and make both the data and the code accessible. --- ### Modular Jets for Supervised Pipelines: Diagnosing Mirage vs Identifiability Authors: Suman Sanyal Published: 2025-12-05 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2512.05638v1 Classical supervised learning evaluates models primarily via predictive risk on hold-out data. Such evaluations quantify how well a function behaves on a distribution, but they do not address whether the internal decomposition of a model is uniquely determined by the data and evaluation design. In this paper, we introduce \emph{Modular Jets} for regression and classification pipelines. Given a task manifold (input space), a modular decomposition, and access to module-level representations, we estimate empirical jets, which are local linear response maps that describe how each module reacts to small structured perturbations of the input. We propose an empirical notion of \emph{mirage} regimes, where multiple distinct modular decompositions induce indistinguishable jets and thus remain observationally equivalent, and contrast this with an \emph{identifiable} regime, where the observed jets single out a decomposition up to natural symmetries. In the setting of two-module linear regression pipelines we prove a jet-identifiability theorem. Under mild rank assumptions and access to module-level jets, the internal factorisation is uniquely determined, whereas risk-only evaluation admits a large family of mirage decompositions that implement the same input-to-output map. We then present an algorithm (MoJet) for empirical jet estimation and mirage diagnostics, and illustrate the framework using linear and deep regression as well as pipeline classification. --- ### Experts-Guided Unbalanced Optimal Transport for ISP Learning from Unpaired and/or Paired Data Authors: Georgy Perevozchikov, Nancy Mehta, Egor Ershov, Radu Timofte Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05635v1 Learned Image Signal Processing (ISP) pipelines offer powerful end-to-end performance but are critically dependent on large-scale paired raw-to-sRGB datasets. This reliance on costly-to-acquire paired data remains a significant bottleneck. To address this challenge, we introduce a novel, unsupervised training framework based on Optimal Transport capable of training arbitrary ISP architectures in both unpaired and paired modes. We are the first to successfully apply Unbalanced Optimal Transport (UOT) for this complex, cross-domain translation task. Our UOT-based framework provides robustness to outliers in the target sRGB data, allowing it to discount atypical samples that would be prohibitively costly to map. A key component of our framework is a novel ``committee of expert discriminators,'' a hybrid adversarial regularizer. This committee guides the optimal transport mapping by providing specialized, targeted gradients to correct specific ISP failure modes, including color fidelity, structural artifacts, and frequency-domain realism. To demonstrate the superiority of our approach, we retrained existing state-of-the-art ISP architectures using our paired and unpaired setups. Our experiments show that while our framework, when trained in paired mode, exceeds the performance of the original paired methods across all metrics, our unpaired mode concurrently achieves quantitative and qualitative performance that rivals, and in some cases surpasses, the original paired-trained counterparts. The code and pre-trained models are available at: https://github.com/gosha20777/EGUOT-ISP.git. --- ### Bounded Graph Clustering with Graph Neural Networks Authors: Kibidi Neocosmos, Diego Baptista, Nicole Ludwig Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05623v2 In community detection, many methods require the user to specify the number of clusters in advance since an exhaustive search over all possible values is computationally infeasible. While some classical algorithms can infer this number directly from the data, this is typically not the case for graph neural networks (GNNs): even when a desired number of clusters is specified, standard GNN-based methods often fail to return the exact number due to the way they are designed. In this work, we address this limitation by introducing a flexible and principled way to control the number of communities discovered by GNNs. Rather than assuming the true number of clusters is known, we propose a framework that allows the user to specify a plausible range and enforce these bounds during training. However, if the user wants an exact number of clusters, it may also be specified and reliably returned. --- ### Hyperparameter Transfer Enables Consistent Gains of Matrix-Preconditioned Optimizers Across Scales Authors: Shikai Qiu, Zixi Chen, Hoang Phan, Qi Lei, Andrew Gordon Wilson Published: 2025-12-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.05620v2 Several recently introduced deep learning optimizers utilizing matrix-level preconditioning have shown promising speedups relative to the current dominant optimizer AdamW, particularly in relatively small-scale experiments. However, efforts to validate and replicate their successes have reported mixed results. To better understand the effectiveness of these optimizers at scale, in this work we investigate how to scale preconditioned optimizers via hyperparameter transfer, building on prior works such as $μ$P. We study how the optimal learning rate and weight decay should scale with model width and depth for a wide range of optimizers, including Shampoo, SOAP, and Muon, accounting for the impact of commonly used techniques such as blocking and grafting. We find that scaling the learning rate according to $μ$P improves transfer, but can still suffer from significant finite-width deviations that cause drifting optimal learning rates, which we show can be mitigated by blocking and explicit spectral normalization. For compute-optimal scaling, we find scaling independent weight decay as $1/\mathrm{width}$ is nearly optimal across optimizers. Applying these scaling rules, we show Muon, SOAP and Shampoo consistently achieve near $1.4\times$ speedup over AdamW for training Llama-architecture language models of sizes ranging from $190$M to $1.4$B, whereas the speedup vanishes rapidly with scale under incorrect scaling. Based on these results and further ablations, we argue that studying optimal hyperparameter transfer is essential for reliably comparing optimizers at scale given a realistic tuning budget. --- ### Enhancing Local Search for MaxSAT with Deep Differentiation Clause Weighting Authors: Menghua Jiang, Haokai Gao, Shuhao Chen, Yin Chen Published: 2025-12-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.05619v1 Partial Maximum Satisfiability (PMS) and Weighted Partial Maximum Satisfiability (WPMS) generalize Maximum Satisfiability (MaxSAT), with broad real-world applications. Recent advances in Stochastic Local Search (SLS) algorithms for solving (W)PMS have mainly focused on designing clause weighting schemes. However, existing methods often fail to adequately distinguish between PMS and WPMS, typically employing uniform update strategies for clause weights and overlooking critical structural differences between the two problem types. In this work, we present a novel clause weighting scheme that, for the first time, updates the clause weights of PMS and WPMS instances according to distinct conditions. This scheme also introduces a new initialization method, which better accommodates the unique characteristics of both instance types. Furthermore, we propose a decimation method that prioritizes satisfying unit and hard clauses, effectively complementing our proposed clause weighting scheme. Building on these methods, we develop a new SLS solver for (W)PMS named DeepDist. Experimental results on benchmarks from the anytime tracks of recent MaxSAT Evaluations show that DeepDist outperforms state-of-the-art SLS solvers. Notably, a hybrid solver combining DeepDist with TT-Open-WBO-Inc surpasses the performance of the MaxSAT Evaluation 2024 winners, SPB-MaxSAT-c-Band and SPB-MaxSAT-c-FPS, highlighting the effectiveness of our approach. The code is available at https://github.com/jmhmaxsat/DeepDist --- ### DistillFSS: Synthesizing Few-Shot Knowledge into a Lightweight Segmentation Model Authors: Pasquale De Marinis, Pieter M. Blok, Uzay Kaymak, Rogier Brussee, Gennaro Vessio, Giovanna Castellano Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05613v1 Cross-Domain Few-Shot Semantic Segmentation (CD-FSS) seeks to segment unknown classes in unseen domains using only a few annotated examples. This setting is inherently challenging: source and target domains exhibit substantial distribution shifts, label spaces are disjoint, and support images are scarce--making standard episodic methods unreliable and computationally demanding at test time. To address these constraints, we propose DistillFSS, a framework that embeds support-set knowledge directly into a model's parameters through a teacher--student distillation process. By internalizing few-shot reasoning into a dedicated layer within the student network, DistillFSS eliminates the need for support images at test time, enabling fast, lightweight inference, while allowing efficient extension to novel classes in unseen domains through rapid teacher-driven specialization. Combined with fine-tuning, the approach scales efficiently to large support sets and significantly reduces computational overhead. To evaluate the framework under realistic conditions, we introduce a new CD-FSS benchmark spanning medical imaging, industrial inspection, and remote sensing, with disjoint label spaces and variable support sizes. Experiments show that DistillFSS matches or surpasses state-of-the-art baselines, particularly in multi-class and multi-shot scenarios, while offering substantial efficiency gains. The code is available at https://github.com/pasqualedem/DistillFSS. --- ### Design-marginal calibration of Gaussian process predictive distributions: Bayesian and conformal approaches Authors: Aurélien Pion, Emmanuel Vazquez Published: 2025-12-05 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2512.05611v1 We study the calibration of Gaussian process (GP) predictive distributions in the interpolation setting from a design-marginal perspective. Conditioning on the data and averaging over a design measure μ, we formalize μ-coverage for central intervals and μ-probabilistic calibration through randomized probability integral transforms. We introduce two methods. cps-gp adapts conformal predictive systems to GP interpolation using standardized leave-one-out residuals, yielding stepwise predictive distributions with finite-sample marginal calibration. bcr-gp retains the GP posterior mean and replaces the Gaussian residual by a generalized normal model fitted to cross-validated standardized residuals. A Bayesian selection rule-based either on a posterior upper quantile of the variance for conservative prediction or on a cross-posterior Kolmogorov-Smirnov criterion for probabilistic calibration-controls dispersion and tail behavior while producing smooth predictive distributions suitable for sequential design. Numerical experiments on benchmark functions compare cps-gp, bcr-gp, Jackknife+ for GPs, and the full conformal Gaussian process, using calibration metrics (coverage, Kolmogorov-Smirnov, integral absolute error) and accuracy or sharpness through the scaled continuous ranked probability score. --- ### NormalView: tree species classification from backpack and aerial lidar data using geometric projections Authors: Juho Korkeala, Jesse Muhojoki, Josef Taher, Klaara Salolahti, Matti Hyyppä, Antero Kukko, Juha Hyyppä Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05610v2 Laser scanning has proven to be an invaluable tool in assessing the decomposition of forest environments. Mobile laser scanning (MLS) has shown to be highly promising for extremely accurate, tree level inventory. In this study, we present NormalView, a projection-based deep learning method for classifying tree species from point cloud data. NormalView embeds local geometric information into two-dimensional projections, in the form of normal vector estimates, and uses the projections as inputs to an image classification network, YOLOv11. In addition, we inspected the effect of multispectral radiometric intensity information on classification performance. We trained and tested our model on high-density MLS data (7 species, ~5000 pts/m2), as well as high-density airborne laser scanning (ALS) data (9 species, >1000 pts/m2). On the MLS data, NormalView achieves an overall accuracy (macro-average accuracy) of 95.5 % (94.8 %), and 91.8 % (79.1 %) on the ALS data. We found that having intensity information from multiple scanners provides benefits in tree species classification, and the best model on the multispectral ALS dataset was a model using intensity information from all three channels of the multispectral ALS. This study demonstrates that projection-based methods, when enhanced with geometric information and coupled with state-of-the-art image classification backbones, can achieve exceptional results. Crucially, these methods rely only on geometric information, and thus are compatible with most sensors. Additionally, we publically release the MLS dataset used in the study, containing 1915 samples. --- ### Fast SceneScript: Fast and Accurate Language-Based 3D Scene Understanding via Multi-Token Prediction Authors: Ruihong Yin, Xuepeng Shi, Oleksandr Bailo, Marco Manfredi, Theo Gevers Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05597v3 Recent perception-generalist approaches based on language models have achieved state-of-the-art results across diverse tasks, including 3D scene layout estimation and 3D object detection, via unified architecture and interface. However, these approaches rely on autoregressive next-token prediction, which is inherently slow. In this work, we introduce Fast SceneScript, a novel structured language model for accurate and efficient 3D scene understanding. Our method employs multi-token prediction (MTP) to reduce the number of autoregressive iterations and significantly accelerate inference. While MTP improves speed, unreliable token predictions can significantly reduce accuracy. To filter out unreliable tokens, we adapt self-speculative decoding (SSD) for structured language models and introduce confidence-guided decoding (CGD) with an improved scoring mechanism for token reliability. Furthermore, we design a parameter-efficient mechanism that reduces the parameter overhead of MTP. Extensive experiments on synthetic and real-world benchmarks demonstrate that Fast SceneScript can generate up to 9 tokens per decoder inference step without compromising accuracy, while adding only $\sim7.5\%$ additional parameters. --- ### Ontology Learning with LLMs: A Benchmark Study on Axiom Identification Authors: Roos M. Bakker, Daan L. Di Scala, Maaike H. T. de Boer, Stephan A. Raaijmakers Published: 2025-12-05 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2512.05594v1 Ontologies are an important tool for structuring domain knowledge, but their development is a complex task that requires significant modelling and domain expertise. Ontology learning, aimed at automating this process, has seen advancements in the past decade with the improvement of Natural Language Processing techniques, and especially with the recent growth of Large Language Models (LLMs). This paper investigates the challenge of identifying axioms: fundamental ontology components that define logical relations between classes and properties. In this work, we introduce an Ontology Axiom Benchmark OntoAxiom, and systematically test LLMs on that benchmark for axiom identification, evaluating different prompting strategies, ontologies, and axiom types. The benchmark consists of nine medium-sized ontologies with together 17.118 triples, and 2.771 axioms. We focus on subclass, disjoint, subproperty, domain, and range axioms. To evaluate LLM performance, we compare twelve LLMs with three shot settings and two prompting strategies: a Direct approach where we query all axioms at once, versus an Axiom-by-Axiom (AbA) approach, where each prompt queries for one axiom only. Our findings show that the AbA prompting leads to higher F1 scores than the direct approach. However, performance varies across axioms, suggesting that certain axioms are more challenging to identify. The domain also influences performance: the FOAF ontology achieves a score of 0.642 for the subclass axiom, while the music ontology reaches only 0.218. Larger LLMs outperform smaller ones, but smaller models may still be viable for resource-constrained settings. Although performance overall is not high enough to fully automate axiom identification, LLMs can provide valuable candidate axioms to support ontology engineers with the development and refinement of ontologies. --- ### Learning High-Fidelity Cloth Animation via Skinning-Free Image Transfer Authors: Rong Wang, Wei Mao, Changsheng Lu, Hongdong Li Published: 2025-12-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.05593v1 We present a novel method for generating 3D garment deformations from given body poses, which is key to a wide range of applications, including virtual try-on and extended reality. To simplify the cloth dynamics, existing methods mostly rely on linear blend skinning to obtain low-frequency posed garment shape and only regress high-frequency wrinkles. However, due to the lack of explicit skinning supervision, such skinning-based approach often produces misaligned shapes when posing the garment, consequently corrupts the high-frequency signals and fails to recover high-fidelity wrinkles. To tackle this issue, we propose a skinning-free approach by independently estimating posed (i) vertex position for low-frequency posed garment shape, and (ii) vertex normal for high-frequency local wrinkle details. In this way, each frequency modality can be effectively decoupled and directly supervised by the geometry of the deformed garment. To further improve the visual quality of animation, we propose to encode both vertex attributes as rendered texture images, so that 3D garment deformation can be equivalently achieved via 2D image transfer. This enables us to leverage powerful pretrained image models to recover fine-grained visual details in wrinkles, while maintaining superior scalability for garments of diverse topologies without relying on manual UV partition. Finally, we propose a multimodal fusion to incorporate constraints from both frequency modalities and robustly recover deformed 3D garments from transferred images. Extensive experiments show that our method significantly improves animation quality on various garment types and recovers finer wrinkles than state-of-the-art methods. --- ### Entropy Ratio Clipping as a Soft Global Constraint for Stable Reinforcement Learning Authors: Zhenpeng Su, Leiyu Pan, Minxuan Lv, Tiehua Mei, Zijia Lin, Yuntao Li, Wenping Hu, Ruiming Tang, Kun Gai, Guorui Zhou Published: 2025-12-05 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2512.05591v2 Large language model post-training relies on reinforcement learning to improve model capability and alignment quality. However, the off-policy training paradigm introduces distribution shift, which often pushes the policy beyond the trust region, leading to training instabilities manifested as fluctuations in policy entropy and unstable gradients. Although PPO-Clip mitigates this issue through importance clipping, it still overlooks the global distributional shift of actions. To address these challenges, we propose using the entropy ratio between the current and previous policies as a new global metric that effectively quantifies the relative change in policy exploration throughout updates. Building on this metric, we introduce an \textbf{Entropy Ratio Clipping} (ERC) mechanism that imposes bidirectional constraints on the entropy ratio. This stabilizes policy updates at the global distribution level and compensates for the inability of PPO-clip to regulate probability shifts of un-sampled actions. We integrate ERC into both DAPO and GPPO reinforcement learning algorithms. Experiments across multiple benchmarks show that ERC consistently improves performance. --- ### Structured Reasoning with Tree-of-Thoughts for Bengali Math Word Problems Authors: Aurprita Mahmood, Sabrin alam, Neloy kumer Sagor, Md. Abdul Hadi, Md. Sehab Al Islam, Minhajul Islam Published: 2025-12-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.05580v1 Mathematical Word Problems (MWPs) are among the most challenging tasks in natural language processing because they require both linguistic understanding and multi-step numerical reasoning. While Chain-of-Thought (CoT) prompting has shown promise, its linear structure often propagates errors, limiting overall effectiveness. To address this limitation, we present the a systematic study of Tree-of-Thought (ToT) reasoning for Bengali MWPs using the SOMADHAN dataset. Owing to computational and token-cost constraints, we evaluate a curated set of 100 representative problems across multiple large language models (LLMs), including GPT-OSS and LLaMA variants, under standard prompting, CoT, and ToT strategies. Our results show that CoT improves baseline accuracy from 78% (standard prompting) to 83% on average, while ToT further increases performance by up to 5 percentage points, achieving 88% accuracy with GPT-OSS-120B. These improvements highlight that ToT is particularly effective in medium-to-large-scale models but may offer less advantage for smaller ones. Overall, our findings establish ToT as a robust framework for solving mathematical problems in low-resource languages such as Bengali. More broadly, this study shows that structured reasoning methods like ToT can provide more reliable and globally consistent outcomes than CoT, paving the way for better reasoning strategies in multilingual NLP. --- ### A Comprehensive Framework for Automated Quality Control in the Automotive Industry Authors: Panagiota Moraiti, Panagiotis Giannikos, Athanasios Mastrogeorgiou, Panagiotis Mavridis, Linghao Zhou, Panagiotis Chatzakos Published: 2025-12-05 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2512.05579v1 This paper presents a cutting-edge robotic inspection solution designed to automate quality control in automotive manufacturing. The system integrates a pair of collaborative robots, each equipped with a high-resolution camera-based vision system to accurately detect and localize surface and thread defects in aluminum high-pressure die casting (HPDC) automotive components. In addition, specialized lenses and optimized lighting configurations are employed to ensure consistent and high-quality image acquisition. The YOLO11n deep learning model is utilized, incorporating additional enhancements such as image slicing, ensemble learning, and bounding-box merging to significantly improve performance and minimize false detections. Furthermore, image processing techniques are applied to estimate the extent of the detected defects. Experimental results demonstrate real-time performance with high accuracy across a wide variety of defects, while minimizing false detections. The proposed solution is promising and highly scalable, providing the flexibility to adapt to various production environments and meet the evolving demands of the automotive industry. --- ### CureAgent: A Training-Free Executor-Analyst Framework for Clinical Reasoning Authors: Ting-Ting Xie, Yixin Zhang Published: 2025-12-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.05576v1 Current clinical agent built on small LLMs, such as TxAgent suffer from a \textit{Context Utilization Failure}, where models successfully retrieve biomedical evidence due to supervised finetuning but fail to ground their diagnosis in that information. In this work, we propose the Executor-Analyst Framework, a modular architecture that decouples the syntactic precision of tool execution from the semantic robustness of clinical reasoning. By orchestrating specialized TxAgents (Executors) with long-context foundation models (Analysts), we mitigate the reasoning deficits observed in monolithic models. Beyond simple modularity, we demonstrate that a Stratified Ensemble strategy significantly outperforms global pooling by preserving evidentiary diversity, effectively addressing the information bottleneck. Furthermore, our stress tests reveal critical scaling insights: (1) a \textit{Context-Performance Paradox}, where extending reasoning contexts beyond 12k tokens introduces noise that degrades accuracy; and (2) the \textit{Curse of Dimensionality} in action spaces, where expanding toolsets necessitates hierarchical retrieval strategies. Crucially, our approach underscores the potential of training-free architectural engineering, achieving state-of-the-art performance on CURE-Bench without the need for expensive end-to-end finetuning. This provides a scalable, agile foundation for the next generation of trustworthy AI-driven therapeutics. Code has been released on https://github.com/June01/CureAgent. ---