# Research Papers — 2025-12 · bundle page 52 of 175 > Full-text markdown of papers #2551–2600 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/52 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/51.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/53.md) ### TraCeR: Transformer-Based Competing Risk Analysis with Longitudinal Covariates Authors: Maxmillan Ries, Sohan Seth Published: 2025-12-19 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2512.18129v2 Survival analysis is a critical tool for modeling time-to-event data. Recent deep learning-based models have reduced various modeling assumptions including proportional hazard and linearity. However, a persistent challenge remains in incorporating longitudinal covariates, with prior work largely focusing on cross-sectional features, and in assessing calibration of these models, with research primarily focusing on discrimination during evaluation. We introduce TraCeR, a transformer-based survival analysis framework for incorporating longitudinal covariates. Based on a factorized self-attention architecture, TraCeR estimates the hazard function from a sequence of measurements, naturally capturing temporal covariate interactions without assumptions about the underlying data-generating process. The framework is inherently designed to handle censored data and competing events. Experiments on multiple real-world datasets demonstrate that TraCeR achieves substantial and statistically significant performance improvements over state-of-the-art methods. Furthermore, our evaluation extends beyond discrimination metrics and assesses model calibration, addressing a key oversight in literature. --- ### SERA-H: Super-Resolution of Sentinel Time Series for Fine-Scale Canopy Height Mapping Authors: Thomas Boudras, Martin Schwartz, Rasmus Fensholt, Martin Brandt, Ibrahim Fayad, Jean-Pierre Wigneron, Gabriel Belouze, Fajwel Fogel, Philippe Ciais Published: 2025-12-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.18128v4 High-resolution mapping of canopy height is essential for forest management and biodiversity monitoring. Although recent studies have led to the advent of deep learning methods using satellite imagery to predict height maps, these approaches often face a trade-off between data accessibility and spatial resolution. To overcome these limitations, we present SERA-H, an end-to-end model combining a super-resolution module (EDSR) and temporal attention encoding (UTAE). Trained under the supervision of high-density LiDAR-derived Canopy Height Models (CHM), our model generates 2.5 m resolution height maps from freely available Sentinel-1 and Sentinel-2 (10 m) time series data. Evaluated against an open-source benchmark dataset in France, SERA-H, with a MAE of 2.6 m and R2 of 0.82, not only outperforms standard Sentinel-1/2 baselines but also approaches the accuracy of methods based on commercial very high-resolution imagery (e.g., SPOT-6/7), while relying solely on freely available data. These results demonstrate that combining high-resolution ALS supervision with the spatio-temporal information embedded in Sentinel time series enables the reconstruction of spatial detail finer than the native resolution of the input imagery. By approaching the accuracy of costly commercial imagery, SERA-H opens the possibility of mapping temperate forests using publicly available data with high revisit frequency. --- ### Efficient Mixture-of-Agents Serving via Tree-Structured Routing, Adaptive Pruning, and Dependency-Aware Prefill-Decode Overlap Authors: Zijun Wang, Yijiahao Qi, Hanqiu Chen, Zishen Wan, Gongjin Sun, Dongyang Li, Shuyi Pei, Cong Hao Published: 2025-12-19 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2512.18126v1 Mixture-of-Agents (MoA) inference can suffer from dense inter-agent communication and low hardware utilization, which jointly inflate serving latency. We present a serving design that targets these bottlenecks through an algorithm-system co-design. First, we replace dense agent interaction graphs with a hierarchical tree topology that induces structured sparsity in inter-agent communication. Second, we introduce a runtime adaptive mechanism that selectively terminates or skips downstream agent invocations using semantic agreement and confidence signals from intermediate outputs. Third, we pipeline agent execution by overlapping incremental prefilling with decoding across dependency-related agents, improving utilization and reducing inference latency. Across representative tasks, this approach substantially reduces end-to-end latency (up to 90%) while maintaining comparable accuracy (within $\pm$1%) relative to dense-connectivity MoA baselines, and can improve accuracy in certain settings. --- ### Exploring polymer classification with a hybrid single-photon quantum approach Authors: Alexandrina Stoyanova, Bogdan Penkovsky Published: 2025-12-19 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2512.18125v1 Polymers exhibit complex architectures and diverse properties that place them at the center of contemporary research in chemistry and materials science. As conventional computational techniques, even multi-scale ones, struggle to capture this complexity, quantum computing offers a promising alternative framework for extracting structure-property relationships. Noisy Intermediate-Scale Quantum (NISQ) devices are commonly used to explore the implementation of algorithms, including quantum neural networks for classification tasks, despite ongoing debate regarding their practical impact. We present a hybrid classical-quantum formalism that couples a classical deep neural network for polymer featurization with a single-photon-based quantum classifier native to photonic quantum computing. This pipeline successfully classifies polymer species by their optical gap, with performance in line between CPU-based noisy simulations and a proof-of-principle run on Quandela's Ascella quantum processor. These findings demonstrate the effectiveness of the proposed computational workflow and indicate that chemistryfrelated classification tasks can already be tackled under the constraints of today's NISQ devices. --- ### Learning Generalizable Neural Operators for Inverse Problems Authors: Adam J. Thorpe, Stepan Tretiakov, Dibakar Roy Sarkar, Krishna Kumar, Ufuk Topcu Published: 2025-12-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.18120v1 Inverse problems challenge existing neural operator architectures because ill-posed inverse maps violate continuity, uniqueness, and stability assumptions. We introduce B2B${}^{-1}$, an inverse basis-to-basis neural operator framework that addresses this limitation. Our key innovation is to decouple function representation from the inverse map. We learn neural basis functions for the input and output spaces, then train inverse models that operate on the resulting coefficient space. This structure allows us to learn deterministic, invertible, and probabilistic models within a single framework, and to choose models based on the degree of ill-posedness. We evaluate our approach on six inverse PDE benchmarks, including two novel datasets, and compare against existing invertible neural operator baselines. We learn probabilistic models that capture uncertainty and input variability, and remain robust to measurement noise due to implicit denoising in the coefficient calculation. Our results show consistent re-simulation performance across varying levels of ill-posedness. By separating representation from inversion, our framework enables scalable surrogate models for inverse problems that generalize across instances, domains, and degrees of ill-posedness. --- ### Distributed Asymmetric Allocation: A Topic Model for Large Imbalanced Corpora in Social Sciences Authors: Kohei Watanabe Published: 2025-12-19 Categories: stat.ME, cs.CL arXiv: https://arxiv.org/abs/2512.18119v1 Social scientists employ latent Dirichlet allocation (LDA) to find highly specific topics in large corpora, but they often struggle in this task because (1) LDA, in general, takes a significant amount of time to fit on large corpora; (2) unsupervised LDA fragments topics into sub-topics in short documents; (3) semi-supervised LDA fails to identify specific topics defined using seed words. To solve these problems, I have developed a new topic model called distributed asymmetric allocation (DAA) that integrates multiple algorithms for efficiently identifying sentences about important topics in large corpora. I evaluate the ability of DAA to identify politically important topics by fitting it to the transcripts of speeches at the United Nations General Assembly between 1991 and 2017. The results show that DAA can classify sentences significantly more accurately and quickly than LDA thanks to the new algorithms. More generally, the results demonstrate that it is important for social scientists to optimize Dirichlet priors of LDA to perform content analysis accurately. --- ### Layout-Aware Text Editing for Efficient Transformation of Academic PDFs to Markdown Authors: Changxu Duan Published: 2025-12-19 Categories: cs.MM, cs.CL, cs.CV, cs.DL arXiv: https://arxiv.org/abs/2512.18115v1 Academic documents stored in PDF format can be transformed into plain text structured markup languages to enhance accessibility and enable scalable digital library workflows. Markup languages allow for easier updates and customization, making academic content more adaptable and accessible to diverse usage, such as linguistic corpus compilation. Such documents, typically delivered in PDF format, contain complex elements including mathematical formulas, figures, headers, and tables, as well as densely layouted text. Existing end-to-end decoder transformer models can transform screenshots of documents into markup language. However, these models exhibit significant inefficiencies; their token-by-token decoding from scratch wastes a lot of inference steps in regenerating dense text that could be directly copied from PDF files. To solve this problem, we introduce EditTrans, a hybrid editing-generation model whose features allow identifying a queue of to-be-edited text from a PDF before starting to generate markup language. EditTrans contains a lightweight classifier fine-tuned from a Document Layout Analysis model on 162,127 pages of documents from arXiv. In our evaluations, EditTrans reduced the transformation latency up to 44.5% compared to end-to-end decoder transformer models, while maintaining transformation quality. Our code and reproducible dataset production scripts are open-sourced. --- ### Microstructure-based Variational Neural Networks for Robust Uncertainty Quantification in Materials Digital Twins Authors: Andreas E. Robertson, Samuel B. Inman, Ashley T. Lenau, Ricardo A. Lebensohn, Dongil Shin, Brad L. Boyce, Remi M. Dingreville Published: 2025-12-19 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2512.18104v1 Aleatoric uncertainties - irremovable variability in microstructure morphology, constituent behavior, and processing conditions - pose a major challenge to developing uncertainty-robust digital twins. We introduce the Variational Deep Material Network (VDMN), a physics-informed surrogate model that enables efficient and probabilistic forward and inverse predictions of material behavior. The VDMN captures microstructure-induced variability by embedding variational distributions within its hierarchical, mechanistic architecture. Using an analytic propagation scheme based on Taylor-series expansion and automatic differentiation, the VDMN efficiently propagates uncertainty through the network during training and prediction. We demonstrate its capabilities in two digital-twin-driven applications: (1) as an uncertainty-aware materials digital twin, it predicts and experimentally validates the nonlinear mechanical variability in additively manufactured polymer composites; and (2) as an inverse calibration engine, it disentangles and quantitatively identifies overlapping sources of uncertainty in constituent properties. Together, these results establish the VDMN as a foundation for uncertainty-robust materials digital twins. --- ### From Coverage to Causes: Data-Centric Fuzzing for JavaScript Engines Authors: Kishan Kumar Ganguly, Tim Menzies Published: 2025-12-19 Categories: cs.SE, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2512.18102v1 Context: Exhaustive fuzzing of modern JavaScript engines is infeasible due to the vast number of program states and execution paths. Coverage-guided fuzzers waste effort on low-risk inputs, often ignoring vulnerability-triggering ones that do not increase coverage. Existing heuristics proposed to mitigate this require expert effort, are brittle, and hard to adapt. Objective: We propose a data-centric, LLM-boosted alternative that learns from historical vulnerabilities to automatically identify minimal static (code) and dynamic (runtime) features for detecting high-risk inputs. Method: Guided by historical V8 bugs, iterative prompting generated 115 static and 49 dynamic features, with the latter requiring only five trace flags, minimizing instrumentation cost. After feature selection, 41 features remained to train an XGBoost model to predict high-risk inputs during fuzzing. Results: Combining static and dynamic features yields over 85% precision and under 1% false alarms. Only 25% of these features are needed for comparable performance, showing that most of the search space is irrelevant. Conclusion: This work introduces feature-guided fuzzing, an automated data-driven approach that replaces coverage with data-directed inference, guiding fuzzers toward high-risk states for faster, targeted, and reproducible vulnerability discovery. To support open science, all scripts and data are available at https://github.com/KKGanguly/DataCentricFuzzJS . --- ### SAM Audio: Segment Anything in Audio Authors: Bowen Shi, Andros Tjandra, John Hoffman, Helin Wang, Yi-Chiao Wu, Luya Gao, Julius Richter, Matt Le, Apoorv Vyas, Sanyuan Chen, Christoph Feichtenhofer, Piotr Dollár et al. Published: 2025-12-19 Categories: eess.AS, cs.CV arXiv: https://arxiv.org/abs/2512.18099v1 General audio source separation is a key capability for multimodal AI systems that can perceive and reason about sound. Despite substantial progress in recent years, existing separation models are either domain-specific, designed for fixed categories such as speech or music, or limited in controllability, supporting only a single prompting modality such as text. In this work, we present SAM Audio, a foundation model for general audio separation that unifies text, visual, and temporal span prompting within a single framework. Built on a diffusion transformer architecture, SAM Audio is trained with flow matching on large-scale audio data spanning speech, music, and general sounds, and can flexibly separate target sources described by language, visual masks, or temporal spans. The model achieves state-of-the-art performance across a diverse suite of benchmarks, including general sound, speech, music, and musical instrument separation in both in-the-wild and professionally produced audios, substantially outperforming prior general-purpose and specialized systems. Furthermore, we introduce a new real-world separation benchmark with human-labeled multimodal prompts and a reference-free evaluation model that correlates strongly with human judgment. --- ### Rethinking Multi-Agent Intelligence Through the Lens of Small-World Networks Authors: Boxuan Wang, Zhuoyun Li, Xiaowei Huang, Yi Dong Published: 2025-12-19 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2512.18094v1 Large language models (LLMs) have enabled multi-agent systems (MAS) in which multiple agents argue, critique, and coordinate to solve complex tasks, making communication topology a first-class design choice. Yet most existing LLM-based MAS either adopt fully connected graphs, simple sparse rings, or ad-hoc dynamic selection, with little structural guidance. In this work, we revisit classic theory on small-world (SW) networks and ask: what changes if we treat SW connectivity as a design prior for MAS? We first bridge insights from neuroscience and complex networks to MAS, highlighting how SW structures balance local clustering and long-range integration. Using multi-agent debate (MAD) as a controlled testbed, experiment results show that SW connectivity yields nearly the same accuracy and token cost, while substantially stabilizing consensus trajectories. Building on this, we introduce an uncertainty-guided rewiring scheme for scaling MAS, where long-range shortcuts are added between epistemically divergent agents using LLM-oriented uncertainty signals (e.g., semantic entropy). This yields controllable SW structures that adapt to task difficulty and agent heterogeneity. Finally, we discuss broader implications of SW priors for MAS design, framing them as stabilizers of reasoning, enhancers of robustness, scalable coordinators, and inductive biases for emergent cognitive roles. --- ### Faithful and Stable Neuron Explanations for Trustworthy Mechanistic Interpretability Authors: Ge Yan, Tuomas Oikarinen, Tsui-Wei, Weng Published: 2025-12-19 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.18092v1 Neuron identification is a popular tool in mechanistic interpretability, aiming to uncover the human-interpretable concepts represented by individual neurons in deep networks. While algorithms such as Network Dissection and CLIP-Dissect achieve great empirical success, a rigorous theoretical foundation remains absent, which is crucial to enable trustworthy and reliable explanations. In this work, we observe that neuron identification can be viewed as the inverse process of machine learning, which allows us to derive guarantees for neuron explanations. Based on this insight, we present the first theoretical analysis of two fundamental challenges: (1) Faithfulness: whether the identified concept faithfully represents the neuron's underlying function and (2) Stability: whether the identification results are consistent across probing datasets. We derive generalization bounds for widely used similarity metrics (e.g. accuracy, AUROC, IoU) to guarantee faithfulness, and propose a bootstrap ensemble procedure that quantifies stability along with BE (Bootstrap Explanation) method to generate concept prediction sets with guaranteed coverage probability. Experiments on both synthetic and real data validate our theoretical results and demonstrate the practicality of our method, providing an important step toward trustworthy neuron identification. --- ### Causal Inference as Distribution Adaptation: Optimizing ATE Risk under Propensity Uncertainty Authors: Ashley Zhang Published: 2025-12-19 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2512.18083v1 Standard approaches to causal inference, such as Outcome Regression and Inverse Probability Weighted Regression Adjustment (IPWRA), are typically derived through the lens of missing data imputation and identification theory. In this work, we unify these methods from a Machine Learning perspective, reframing ATE estimation as a \textit{domain adaptation problem under distribution shift}. We demonstrate that the canonical Hajek estimator is a special case of IPWRA restricted to a constant hypothesis class, and that IPWRA itself is fundamentally Importance-Weighted Empirical Risk Minimization designed to correct for the covariate shift between the treated sub-population and the target population. Leveraging this unified framework, we critically examine the optimization objectives of Doubly Robust estimators. We argue that standard methods enforce \textit{sufficient but not necessary} conditions for consistency by requiring outcome models to be individually unbiased. We define the true "ATE Risk Function" and show that minimizing it requires only that the biases of the treated and control models structurally cancel out. Exploiting this insight, we propose the \textbf{Joint Robust Estimator (JRE)}. Instead of treating propensity estimation and outcome modeling as independent stages, JRE utilizes bootstrap-based uncertainty quantification of the propensity score to train outcome models jointly. By optimizing for the expected ATE risk over the distribution of propensity scores, JRE leverages model degrees of freedom to achieve robustness against propensity misspecification. Simulation studies demonstrate that JRE achieves up to a 15\% reduction in MSE compared to standard IPWRA in finite-sample regimes with misspecified outcome models. --- ### Uncertainty-Gated Region-Level Retrieval for Robust Semantic Segmentation Authors: Shreshth Rajan, Raymond Liu Published: 2025-12-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.18082v1 Semantic segmentation of outdoor street scenes plays a key role in applications such as autonomous driving, mobile robotics, and assistive technology for visually-impaired pedestrians. For these applications, accurately distinguishing between key surfaces and objects such as roads, sidewalks, vehicles, and pedestrians is essential for maintaining safety and minimizing risks. Semantic segmentation must be robust to different environments, lighting and weather conditions, and sensor noise, while being performed in real-time. We propose a region-level, uncertainty-gated retrieval mechanism that improves segmentation accuracy and calibration under domain shift. Our best method achieves an 11.3% increase in mean intersection-over-union while reducing retrieval cost by 87.5%, retrieving for only 12.5% of regions compared to 100% for always-on baseline. --- ### From Prompt to Product: A Human-Centered Benchmark of Agentic App Generation Systems Authors: Marcos Ortiz, Justin Hill, Collin Overbay, Ingrida Semenec, Frederic Sauve-Hoover, Jim Schwoebel, Joel Shor Published: 2025-12-19 Categories: cs.HC, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2512.18080v2 Agentic AI systems capable of generating full-stack web applications from natural language prompts ("prompt- to-app") represent a significant shift in software development. However, evaluating these systems remains challenging, as visual polish, functional correctness, and user trust are often misaligned. As a result, it is unclear how existing prompt-to-app tools compare under realistic, human-centered evaluation criteria. In this paper, we introduce a human-centered benchmark for evaluating prompt-to-app systems and conduct a large-scale comparative study of three widely used platforms: Replit, Bolt, and Firebase Studio. Using a diverse set of 96 prompts spanning common web application tasks, we generate 288 unique application artifacts. We evaluate these systems through a large-scale human-rater study involving 205 participants and 1,071 quality-filtered pairwise comparisons, assessing task-based ease of use, visual appeal, perceived completeness, and user trust. Our results show that these systems are not interchangeable: Firebase Studio consistently outperforms competing platforms across all human-evaluated dimensions, achieving the highest win rates for ease of use, trust, visual appeal, and visual appropriateness. Bolt performs competitively on visual appeal but trails Firebase on usability and trust, while Replit underperforms relative to both across most metrics. These findings highlight a persistent gap between visual polish and functional reliability in prompt-to-app systems and demonstrate the necessity of interactive, task-based evaluation. We release our benchmark framework, prompt set, and generated artifacts to support reproducible evaluation and future research in agentic application generation. --- ### Characterising Behavioural Families and Dynamics of Promotional Twitter Bots via Sequence-Based Modelling Authors: Ohoud Alzahrani, Russell Beale, Robert J. Hendley Published: 2025-12-19 Categories: cs.SI, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2512.18077v1 This paper asks whether promotional Twitter/X bots form behavioural families and whether members evolve similarly. We analyse 2,798,672 tweets from 2,615 ground-truth promotional bot accounts (2006-2021), focusing on complete years 2009 to 2020. Each bot is encoded as a sequence of symbolic blocks (``digital DNA'') from seven categorical post-level behavioural features (posting action, URL, media, text duplication, hashtags, emojis, sentiment), preserving temporal order only. Using non-overlapping blocks (k=7), cosine similarity over block-frequency vectors, and hierarchical clustering, we obtain four coherent families: Unique Tweeters, Duplicators with URLs, Content Multipliers, and Informed Contributors. Families share behavioural cores but differ systematically in engagement strategies and life-cycle dynamics (beginning/middle/end). We then model behavioural change as mutations. Within each family we align sequences via multiple sequence alignment (MSA) and label events as insertions, deletions, substitutions, alterations, and identity. This quantifies mutation rates, change-prone blocks/features, and mutation hotspots. Deletions and substitutions dominate, insertions are rare, and mutation profiles differ by family, with hotspots early for some families and dispersed for others. Finally, we test predictive value: bots within the same family share mutations more often than bots across families; closer bots share and propagate mutations more than distant ones; and responses to external triggers (e.g., Christmas, Halloween) follow family-specific, partly predictable patterns. Overall, sequence-based family modelling plus mutation analysis provides a fine-grained account of how promotional bot behaviour adapts over time. --- ### FPBench: A Comprehensive Benchmark of Multimodal Large Language Models for Fingerprint Analysis Authors: Ekta Gavas, Sudipta Banerjee, Chinmay Hegde, Nasir Memon Published: 2025-12-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.18073v2 Multimodal LLMs (MLLMs) are capable of performing complex data analysis, visual question answering, generation, and reasoning tasks. However, their ability to analyze biometric data is relatively underexplored. In this work, we investigate the effectiveness of MLLMs in understanding fine structural and textural details present in fingerprint images. To this end, we design a comprehensive benchmark, FPBench, to evaluate 20 MLLMs (open-source and proprietary models) across 7 real and synthetic datasets on a suite of 8 biometric and forensic tasks (e.g., pattern analysis, fingerprint verification, real versus synthetic classification, etc.) using zero-shot and chain-of-thought prompting strategies. We further fine-tune vision and language encoders on a subset of open-source MLLMs to demonstrate domain adaptation. FPBench is a novel benchmark designed as a first step towards developing foundation models in fingerprints. Our findings indicate fine-tuning of vision and language encoders improves the performance by 7%-39%. Our codes are available at https://github.com/Ektagavas/FPBench. --- ### Statistical laws and linguistics differ in naturalistic video and fictional conversations Authors: Ashley M. A. Fehr, Calla G. Beauregard, Julia Witte Zimmerman, Katie Ekström, Pablo Rosillo-Rodes, Christopher M. Danforth, Peter Sheridan Dodds Published: 2025-12-19 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2512.18072v3 Conversation is a cornerstone of social connection and is linked to well-being outcomes. Conversations vary widely in type with some portion generating complex, dynamic stories. One approach to studying how conversations unfold in time is through statistical patterns such as Heaps' law, which holds that vocabulary size scales with document length. Little work on Heaps' law has looked at conversation and considered how language features impact scaling. We measure Heaps' law for conversations recorded in two distinct mediums: 1. Strangers brought together on video chat and 2. Fictional characters in movies. We find that scaling of vocabulary size differs by parts of speech, suggesting a less efficient purpose in communication by medium. --- ### Graph-based Nearest Neighbors with Dynamic Updates via Random Walks Authors: Nina Mishra, Yonatan Naamad, Tal Wagner, Lichen Zhang Published: 2025-12-19 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2512.18060v1 Approximate nearest neighbor search (ANN) is a common way to retrieve relevant search results, especially now in the context of large language models and retrieval augmented generation. One of the most widely used algorithms for ANN is based on constructing a multi-layer graph over the dataset, called the Hierarchical Navigable Small World (HNSW). While this algorithm supports insertion of new data, it does not support deletion of existing data. Moreover, deletion algorithms described by prior work come at the cost of increased query latency, decreased recall, or prolonged deletion time. In this paper, we propose a new theoretical framework for graph-based ANN based on random walks. We then utilize this framework to analyze a randomized deletion approach that preserves hitting time statistics compared to the graph before deleting the point. We then turn this theoretical framework into a deterministic deletion algorithm, and show that it provides better tradeoff between query latency, recall, deletion time, and memory usage through an extensive collection of experiments. --- ### Approximation and learning with compositional tensor trains Authors: Martin Eigel, Charles Miranda, Anthony Nouy, David Sommer Published: 2025-12-19 Categories: math.NA, cs.LG arXiv: https://arxiv.org/abs/2512.18059v1 We introduce compositional tensor trains (CTTs) for the approximation of multivariate functions, a class of models obtained by composing low-rank functions in the tensor-train format. This format can encode standard approximation tools, such as (sparse) polynomials, deep neural networks (DNNs) with fixed width, or tensor networks with arbitrary permutation of the inputs, or more general affine coordinate transformations, with similar complexities. This format can be viewed as a DNN with width exponential in the input dimension and structured weights matrices. Compared to DNNs, this format enables controlled compression at the layer level using efficient tensor algebra. On the optimization side, we derive a layerwise algorithm inspired by natural gradient descent, allowing to exploit efficient low-rank tensor algebra. This relies on low-rank estimations of Gram matrices, and tensor structured random sketching. Viewing the format as a discrete dynamical system, we also derive an optimization algorithm inspired by numerical methods in optimal control. Numerical experiments on regression tasks demonstrate the expressivity of the new format and the relevance of the proposed optimization algorithms. Overall, CTTs combine the expressivity of compositional models with the algorithmic efficiency of tensor algebra, offering a scalable alternative to standard deep neural networks. --- ### FOODER: Real-time Facial Authentication and Expression Recognition Authors: Sabri Mustafa Kahya, Muhammet Sami Yavuz, Boran Hamdi Sivrikaya, Eckehard Steinbach Published: 2025-12-19 Categories: cs.CV, cs.AI, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2512.18057v1 Out-of-distribution (OOD) detection is essential for the safe deployment of neural networks, as it enables the identification of samples outside the training domain. We present FOODER, a real-time, privacy-preserving radar-based framework that integrates OOD-based facial authentication with facial expression recognition. FOODER operates using low-cost frequency-modulated continuous-wave (FMCW) radar and exploits both range-Doppler and micro range-Doppler representations. The authentication module employs a multi-encoder multi-decoder architecture with Body Part (BP) and Intermediate Linear Encoder-Decoder (ILED) components to classify a single enrolled individual as in-distribution while detecting all other faces as OOD. Upon successful authentication, an expression recognition module is activated. Concatenated radar representations are processed by a ResNet block to distinguish between dynamic and static facial expressions. Based on this categorization, two specialized MobileViT networks are used to classify dynamic expressions (smile, shock) and static expressions (neutral, anger). This hierarchical design enables robust facial authentication and fine-grained expression recognition while preserving user privacy by relying exclusively on radar data. Experiments conducted on a dataset collected with a 60 GHz short-range FMCW radar demonstrate that FOODER achieves an AUROC of 94.13% and an FPR95 of 18.12% for authentication, along with an average expression recognition accuracy of 94.70%. FOODER outperforms state-of-the-art OOD detection methods and several transformer-based architectures while operating efficiently in real time. --- ### Probabilistic Digital Twins of Users: Latent Representation Learning with Statistically Validated Semantics Authors: Daniel David Published: 2025-12-19 Categories: cs.LG, cs.SI arXiv: https://arxiv.org/abs/2512.18056v1 Understanding user identity and behavior is central to applications such as personalization, recommendation, and decision support. Most existing approaches rely on deterministic embeddings or black-box predictive models, offering limited uncertainty quantification and little insight into what latent representations encode. We propose a probabilistic digital twin framework in which each user is modeled as a latent stochastic state that generates observed behavioral data. The digital twin is learned via amortized variational inference, enabling scalable posterior estimation while retaining a fully probabilistic interpretation. We instantiate this framework using a variational autoencoder (VAE) applied to a user-response dataset designed to capture stable aspects of user identity. Beyond standard reconstruction-based evaluation, we introduce a statistically grounded interpretation pipeline that links latent dimensions to observable behavioral patterns. By analyzing users at the extremes of each latent dimension and validating differences using nonparametric hypothesis tests and effect sizes, we demonstrate that specific dimensions correspond to interpretable traits such as opinion strength and decisiveness. Empirically, we find that user structure is predominantly continuous rather than discretely clustered, with weak but meaningful structure emerging along a small number of dominant latent axes. These results suggest that probabilistic digital twins can provide interpretable, uncertainty-aware representations that go beyond deterministic user embeddings. --- ### YolovN-CBi: A Lightweight and Efficient Architecture for Real-Time Detection of Small UAVs Authors: Ami Pandat, Punna Rajasekhar, Gopika Vinod, Rohit Shukla Published: 2025-12-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.18046v3 Unmanned Aerial Vehicles, commonly known as, drones pose increasing risks in civilian and defense settings, demanding accurate and real-time drone detection systems. However, detecting drones is challenging because of their small size, rapid movement, and low visual contrast. A modified architecture of YolovN called the YolovN-CBi is proposed that incorporates the Convolutional Block Attention Module (CBAM) and the Bidirectional Feature Pyramid Network (BiFPN) to improve sensitivity to small object detections. A curated training dataset consisting of 28K images is created with various flying objects and a local test dataset is collected with 2500 images consisting of very small drone objects. The proposed architecture is evaluated on four benchmark datasets, along with the local test dataset. The baseline Yolov5 and the proposed Yolov5-CBi architecture outperform newer Yolo versions, including Yolov8 and Yolov12, in the speed-accuracy trade-off for small object detection. Four other variants of the proposed CBi architecture are also proposed and evaluated, which vary in the placement and usage of CBAM and BiFPN. These variants are further distilled using knowledge distillation techniques for edge deployment, using a Yolov5m-CBi teacher and a Yolov5n-CBi student. The distilled model achieved a mA@P0.5:0.9 of 0.6573, representing a 6.51% improvement over the teacher's score of 0.6171, highlighting the effectiveness of the distillation process. The distilled model is 82.9% faster than the baseline model, making it more suitable for real-time drone detection. These findings highlight the effectiveness of the proposed CBi architecture, together with the distilled lightweight models in advancing efficient and accurate real-time detection of small UAVs. --- ### Securing Agentic AI Systems -- A Multilayer Security Framework Authors: Sunil Arora, John Hastings Published: 2025-12-19 Categories: cs.CR, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2512.18043v1 Securing Agentic Artificial Intelligence (AI) systems requires addressing the complex cyber risks introduced by autonomous, decision-making, and adaptive behaviors. Agentic AI systems are increasingly deployed across industries, organizations, and critical sectors such as cybersecurity, finance, and healthcare. However, their autonomy introduces unique security challenges, including unauthorized actions, adversarial manipulation, and dynamic environmental interactions. Existing AI security frameworks do not adequately address these challenges or the unique nuances of agentic AI. This research develops a lifecycle-aware security framework specifically designed for agentic AI systems using the Design Science Research (DSR) methodology. The paper introduces MAAIS, an agentic security framework, and the agentic AI CIAA (Confidentiality, Integrity, Availability, and Accountability) concept. MAAIS integrates multiple defense layers to maintain CIAA across the AI lifecycle. Framework validation is conducted by mapping with the established MITRE ATLAS (Adversarial Threat Landscape for Artificial-Intelligence Systems) AI tactics. The study contributes a structured, standardized, and framework-based approach for the secure deployment and governance of agentic AI in enterprise environments. This framework is intended for enterprise CISOs, security, AI platform, and engineering teams and offers a detailed step-by-step approach to securing agentic AI workloads. --- ### Narrative Consolidation: Formulating a New Task for Unifying Multi-Perspective Accounts Authors: Roger A. Finger, Eduardo G. Cortes, Sandro J. Rigo, Gabriel de O. Ramos Published: 2025-12-19 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2512.18041v2 Processing overlapping narrative documents, such as legal testimonies or historical accounts, often aims not for compression but for a unified, coherent, and chronologically sound text. Standard Multi-Document Summarization (MDS), with its focus on conciseness, fails to preserve narrative flow. This paper formally defines this challenge as a new NLP task, Narrative Consolidation, focusing on chronological integrity, completeness, and the fusion of complementary details. We establish the resources needed to study it: a formal task definition, an evaluation paradigm, the Gospel Consolidation Language Resource -- a benchmark built from the four Biblical Gospels with 169 canonical events, cross-document alignments, and a reference consolidation -- and a suite of reference systems, ranging from timeline-agnostic heuristics to the Temporal Alignment Event Graph (TAEG). Benchmarking yields three findings. First, the explicit temporal backbone is the dominant factor: every system granted the canonical timeline raises ROUGE-L F1 from 0.206 to at least 0.81. Second, on a fusion-style reference, a simple length heuristic is remarkably strong (0.947 ROUGE-L F1), outperforming graph-based selection (0.846). Third, ablations show the discriminative signal resides in temporal edges, while intra-cluster lexical similarity is uninformative. These results establish Narrative Consolidation as a distinct task and pose an open challenge: to surpass that heuristic with a principled selection mechanism and true fusion of complementary details. --- ### NodMAISI: Nodule-Oriented Medical AI for Synthetic Imaging Authors: Fakrul Islam Tushar, Ehsan Samei, Cynthia Rudin, Joseph Y. Lo Published: 2025-12-19 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.18038v1 Objective: Although medical imaging datasets are increasingly available, abnormal and annotation-intensive findings critical to lung cancer screening, particularly small pulmonary nodules, remain underrepresented and inconsistently curated. Methods: We introduce NodMAISI, an anatomically constrained, nodule-oriented CT synthesis and augmentation framework trained on a unified multi-source cohort (7,042 patients, 8,841 CTs, 14,444 nodules). The framework integrates: (i) a standardized curation and annotation pipeline linking each CT with organ masks and nodule-level annotations, (ii) a ControlNet-conditioned rectified-flow generator built on MAISI-v2's foundational blocks to enforce anatomy- and lesion-consistent synthesis, and (iii) lesion-aware augmentation that perturbs nodule masks (controlled shrinkage) while preserving surrounding anatomy to generate paired CT variants. Results: Across six public test datasets, NodMAISI improved distributional fidelity relative to MAISI-v2 (real-to-synthetic FID range 1.18 to 2.99 vs 1.69 to 5.21). In lesion detectability analysis using a MONAI nodule detector, NodMAISI substantially increased average sensitivity and more closely matched clinical scans (IMD-CT: 0.69 vs 0.39; DLCS24: 0.63 vs 0.20), with the largest gains for sub-centimeter nodules where MAISI-v2 frequently failed to reproduce the conditioned lesion. In downstream nodule-level malignancy classification trained on LUNA25 and externally evaluated on LUNA16, LNDbv4, and DLCS24, NodMAISI augmentation improved AUC by 0.07 to 0.11 at <=20% clinical data and by 0.12 to 0.21 at 10%, consistently narrowing the performance gap under data scarcity. --- ### Towards Benchmarking Privacy Vulnerabilities in Selective Forgetting with Large Language Models Authors: Wei Qian, Chenxu Zhao, Yangyi Li, Mengdi Huai Published: 2025-12-19 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2512.18035v1 The rapid advancements in artificial intelligence (AI) have primarily focused on the process of learning from data to acquire knowledgeable learning systems. As these systems are increasingly deployed in critical areas, ensuring their privacy and alignment with human values is paramount. Recently, selective forgetting (also known as machine unlearning) has shown promise for privacy and data removal tasks, and has emerged as a transformative paradigm shift in the field of AI. It refers to the ability of a model to selectively erase the influence of previously seen data, which is especially important for compliance with modern data protection regulations and for aligning models with human values. Despite its promise, selective forgetting raises significant privacy concerns, especially when the data involved come from sensitive domains. While new unlearning-induced privacy attacks are continuously proposed, each is shown to outperform its predecessors using different experimental settings, which can lead to overly optimistic and potentially unfair assessments that may disproportionately favor one particular attack over the others. In this work, we present the first comprehensive benchmark for evaluating privacy vulnerabilities in selective forgetting. We extensively investigate privacy vulnerabilities of machine unlearning techniques and benchmark privacy leakage across a wide range of victim data, state-of-the-art unlearning privacy attacks, unlearning methods, and model architectures. We systematically evaluate and identify critical factors related to unlearning-induced privacy leakage. With our novel insights, we aim to provide a standardized tool for practitioners seeking to deploy customized unlearning applications with faithful privacy assessments. --- ### Accelerating Discrete Facility Layout Optimization: A Hybrid CDCL and CP-SAT Architecture Authors: Joshua Gibson, Kapil Dhakal Published: 2025-12-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.18034v3 Discrete facility layout design involves placing physical entities to minimize handling costs while adhering to strict safety and spatial constraints. This combinatorial problem is typically addressed using Mixed Integer Linear Programming (MILP) or Constraint Programming (CP), though these methods often face scalability challenges as constraint density increases. This study systematically evaluates the potential of Conflict-Driven Clause Learning (CDCL) with VSIDS heuristics as an alternative computational engine for discrete layout problems. Using a unified benchmarking harness, we conducted a controlled comparison of CDCL, CP-SAT, and MILP across varying grid sizes and constraint densities. Experimental results reveal a distinct performance dichotomy: while CDCL struggles with optimization objectives due to cost-blind branching, it demonstrates unrivaled dominance in feasibility detection, solving highly constrained instances orders of magnitude faster than competing paradigms. Leveraging this finding, we developed a novel "Warm-Start" hybrid architecture that utilizes CDCL to rapidly generate valid feasibility hints, which are then injected into a CP-SAT optimizer. Our results confirm that this layered approach successfully accelerates exact optimization, using SAT-driven pruning to bridge the gap between rapid satisfiability and proven optimality. --- ### A Dataset and Benchmarks for Atrial Fibrillation Detection from Electrocardiograms of Intensive Care Unit Patients Authors: Sarah Nassar, Nooshin Maghsoodi, Sophia Mannina, Shamel Addas, Stephanie Sibley, Gabor Fichtinger, David Pichora, David Maslove, Purang Abolmaesumi, Parvin Mousavi Published: 2025-12-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.18031v2 Objective: Atrial fibrillation (AF) is the most common cardiac arrhythmia experienced by intensive care unit (ICU) patients and can cause adverse health effects. In this study, we publish a labelled ICU dataset and benchmarks for AF detection. Methods: We compared machine learning models across three data-driven artificial intelligence (AI) approaches: feature-based classifiers, deep learning (DL), and ECG foundation models (FMs). This comparison addresses a critical gap in the literature and aims to pinpoint which AI approach is best for accurate AF detection. Electrocardiograms (ECGs) from a Canadian ICU and the 2021 PhysioNet/Computing in Cardiology Challenge were used to conduct the experiments. Multiple training configurations were tested, ranging from zero-shot inference to transfer learning. Results: On average and across both datasets, ECG FMs performed best, followed by DL, then feature-based classifiers. The model that achieved the top F1 score on our ICU test set was ECG-FM through a transfer learning strategy (F1=0.89). Conclusion: This study demonstrates promising potential for using AI to build an automatic patient monitoring system. Significance: By publishing our labelled ICU dataset (LinkToBeAdded) and performance benchmarks, this work enables the research community to continue advancing the state-of-the-art in AF detection in the ICU. https://physionet.org/content/kingston-icu-af/ --- ### Long-range electrostatics for machine learning interatomic potentials is easier than we thought Authors: Dongjin Kim, Bingqing Cheng Published: 2025-12-19 Categories: physics.comp-ph, cond-mat.mtrl-sci, cs.LG, physics.chem-ph arXiv: https://arxiv.org/abs/2512.18029v1 The lack of long-range electrostatics is a key limitation of modern machine learning interatomic potentials (MLIPs), hindering reliable applications to interfaces, charge-transfer reactions, polar and ionic materials, and biomolecules. In this Perspective, we distill two design principles behind the Latent Ewald Summation (LES) framework, which can capture long-range interactions, charges, and electrical response just by learning from standard energy and force training data: (i) use a Coulomb functional form with environment-dependent charges to capture electrostatic interactions, and (ii) avoid explicit training on ambiguous density functional theory (DFT) partial charges. When both principles are satisfied, substantial flexibility remains: essentially any short-range MLIP can be augmented; charge equilibration schemes can be added when desired; dipoles and Born effective charges can be inferred or finetuned; and charge/spin-state embeddings or tensorial targets can be further incorporated. We also discuss current limitations and open challenges. Together, these minimal, physics-guided design rules suggest that incorporating long-range electrostatics into MLIPs is simpler and perhaps more broadly applicable than is commonly assumed. --- ### Embodied4C: Measuring What Matters for Embodied Vision-Language Navigation Authors: Tin Stribor Sohn, Maximilian Dillitzer, Jason J. Corso, Eric Sax Published: 2025-12-19 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2512.18028v1 Vision-language navigation requires agents to reason and act under constraints of embodiment. While vision-language models (VLMs) demonstrate strong generalization, current benchmarks provide limited understanding of how embodiment -- i.e., the choice of physical platform, sensor configuration, and modality alignment -- influences perception, reasoning, and control. We introduce Embodied4C, a closed-loop benchmark designed as a Turing test for embodied reasoning. The benchmark evaluates the core embodied capabilities of VLMs across three heterogeneous embodiments -- autonomous vehicles, aerial drones, and robotic manipulators -- through approximately 1.1K one-shot reasoning questions and 58 goal-directed navigation tasks. These tasks jointly assess four foundational dimensions: semantic, spatial, temporal, and physical reasoning. Each embodiment presents dynamic sensor configurations and environment variations to probe generalization beyond platform-specific adaptation. To prevent embodiment overfitting, Embodied4C integrates domain-far queries targeting abstract and cross-context reasoning. Comprehensive evaluation across ten state-of-the-art VLMs and four embodied control baselines shows that cross-modal alignment and instruction tuning matter more than scale, while spatial and temporal reasoning remains the primary bottleneck for reliable embodied competence. --- ### CoPE: A Small Language Model for Steerable and Scalable Content Labeling Authors: Samidh Chakrabarti, David Willner, Kevin Klyman, Tiffany Saade, Emily Capstick, Sabina Nong Published: 2025-12-19 Categories: cs.CL, cs.CY, cs.SI arXiv: https://arxiv.org/abs/2512.18027v1 This paper details the methodology behind CoPE, a policy-steerable small language model capable of fast and accurate content labeling. We present a novel training curricula called Contradictory Example Training that enables the model to learn policy interpretation rather than mere policy memorization. We also present a novel method for generating content policies, called Binocular Labeling, which enables rapid construction of unambiguous training datasets. When evaluated across seven different harm areas, CoPE exhibits equal or superior accuracy to frontier models at only 1% of their size. We openly release a 9 billion parameter version of the model that can be run on a single consumer-grade GPU. Models like CoPE represent a paradigm shift for classifier systems. By turning an ML task into a policy writing task, CoPE opens up new design possibilities for the governance of online platforms. --- ### Shuttling Compiler for Trapped-Ion Quantum Computers Based on Fine-Tuned Large Language Models Authors: Fabian Kreppel, Reza Salkhordeh, Ferdinand Schmidt-Kaler, André Brinkmann Published: 2025-12-19 Categories: quant-ph, cs.ET, cs.LG arXiv: https://arxiv.org/abs/2512.18021v4 In trapped-ion quantum computers, qubits must be shuttled between segments to interact. The routing logic that schedules these movements is written by hand for every new trap architecture. We present shuttling compilers based on five large language models (LLMs). Each LLM is fine-tuned on shuttling schedules produced by hand-coded heuristics for linear and branched one-dimensional trap architectures. We investigate how the shuttling operation counts of their schedules compare with those of the heuristics and how far they generalize to unseen architectures. For circuits of up to 16 qubits, the fine-tuned LLMs generate valid schedules on both training architectures, more often the fewer qubits a circuit has. In 12% of the compilations yielding a schedule, the best of ten runs needs up to 21% fewer operations than the heuristic baselines, after a rule-based post-processing step. A single run of one fine-tuned LLM produces a valid shuttling schedule for a previously unseen four-way branched architecture. This is preliminary evidence of cross-architecture generalization. On two other unseen architectures no LLM produces a valid schedule. Thus, LLM-learned shuttling compilation is feasible, and we show how far it currently reaches. --- ### Specification and Detection of LLM Code Smells Authors: Brahim Mahmoudi, Zacharie Chenail-Larcher, Naouel Moha, Quentin Stiévenart, Florent Avellaneda Published: 2025-12-19 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2512.18020v1 Large Language Models (LLMs) have gained massive popularity in recent years and are increasingly integrated into software systems for diverse purposes. However, poorly integrating them in source code may undermine software system quality. Yet, to our knowledge, there is no formal catalog of code smells specific to coding practices for LLM inference. In this paper, we introduce the concept of LLM code smells and formalize five recurrent problematic coding practices related to LLM inference in software systems, based on relevant literature. We extend the detection tool SpecDetect4AI to cover the newly defined LLM code smells and use it to validate their prevalence in a dataset of 200 open-source LLM systems. Our results show that LLM code smells affect 60.50% of the analyzed systems, with a detection precision of 86.06%. --- ### ReGal: A First Look at PPO-based Legal AI for Judgment Prediction and Summarization in India Authors: Shubham Kumar Nigam, Tanuj Tyagi, Siddharth Shukla, Aditya Kumar Guru, Balaramamahanthi Deepak Patnaik, Danush Khanna, Noel Shallum, Kripabandhu Ghosh, Arnab Bhattacharya Published: 2025-12-19 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.18014v1 This paper presents an early exploration of reinforcement learning methodologies for legal AI in the Indian context. We introduce Reinforcement Learning-based Legal Reasoning (ReGal), a framework that integrates Multi-Task Instruction Tuning with Reinforcement Learning from AI Feedback (RLAIF) using Proximal Policy Optimization (PPO). Our approach is evaluated across two critical legal tasks: (i) Court Judgment Prediction and Explanation (CJPE), and (ii) Legal Document Summarization. Although the framework underperforms on standard evaluation metrics compared to supervised and proprietary models, it provides valuable insights into the challenges of applying RL to legal texts. These challenges include reward model alignment, legal language complexity, and domain-specific adaptation. Through empirical and qualitative analysis, we demonstrate how RL can be repurposed for high-stakes, long-document tasks in law. Our findings establish a foundation for future work on optimizing legal reasoning pipelines using reinforcement learning, with broader implications for building interpretable and adaptive legal AI systems. --- ### Robotic VLA Benefits from Joint Learning with Motion Image Diffusion Authors: Yu Fang, Kanchana Ranasinghe, Le Xue, Honglu Zhou, Juntao Tan, Ran Xu, Shelby Heinecke, Caiming Xiong, Silvio Savarese, Daniel Szafir, Mingyu Ding, Michael S. Ryoo et al. Published: 2025-12-19 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2512.18007v1 Vision-Language-Action (VLA) models have achieved remarkable progress in robotic manipulation by mapping multimodal observations and instructions directly to actions. However, they typically mimic expert trajectories without predictive motion reasoning, which limits their ability to reason about what actions to take. To address this limitation, we propose joint learning with motion image diffusion, a novel strategy that enhances VLA models with motion reasoning capabilities. Our method extends the VLA architecture with a dual-head design: while the action head predicts action chunks as in vanilla VLAs, an additional motion head, implemented as a Diffusion Transformer (DiT), predicts optical-flow-based motion images that capture future dynamics. The two heads are trained jointly, enabling the shared VLM backbone to learn representations that couple robot control with motion knowledge. This joint learning builds temporally coherent and physically grounded representations without modifying the inference pathway of standard VLAs, thereby maintaining test-time latency. Experiments in both simulation and real-world environments demonstrate that joint learning with motion image diffusion improves the success rate of pi-series VLAs to 97.5% on the LIBERO benchmark and 58.0% on the RoboTwin benchmark, yielding a 23% improvement in real-world performance and validating its effectiveness in enhancing the motion reasoning capability of large-scale VLAs. --- ### Seeing Justice Clearly: Handwritten Legal Document Translation with OCR and Vision-Language Models Authors: Shubham Kumar Nigam, Parjanya Aditya Shukla, Noel Shallum, Arnab Bhattacharya Published: 2025-12-19 Categories: cs.CV, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2512.18004v1 Handwritten text recognition (HTR) and machine translation continue to pose significant challenges, particularly for low-resource languages like Marathi, which lack large digitized corpora and exhibit high variability in handwriting styles. The conventional approach to address this involves a two-stage pipeline: an OCR system extracts text from handwritten images, which is then translated into the target language using a machine translation model. In this work, we explore and compare the performance of traditional OCR-MT pipelines with Vision Large Language Models that aim to unify these stages and directly translate handwritten text images in a single, end-to-end step. Our motivation is grounded in the urgent need for scalable, accurate translation systems to digitize legal records such as FIRs, charge sheets, and witness statements in India's district and high courts. We evaluate both approaches on a curated dataset of handwritten Marathi legal documents, with the goal of enabling efficient legal document processing, even in low-resource environments. Our findings offer actionable insights toward building robust, edge-deployable solutions that enhance access to legal information for non-native speakers and legal professionals alike. --- ### Name That Part: 3D Part Segmentation and Naming Authors: Soumava Paul, Prakhar Kaushik, Ankit Vaidya, Anand Bhattad, Alan Yuille Published: 2025-12-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.18003v2 We address semantic 3D part segmentation: decomposing objects into parts with meaningful names. While datasets exist with part annotations, their definitions are inconsistent across datasets, limiting robust training. Previous methods produce unlabeled decompositions or retrieve single parts without complete shape annotations. We propose ALIGN-Parts, which formulates part naming as a direct set alignment task. Our method decomposes shapes into partlets - implicit 3D part representations - matched to part descriptions via bipartite assignment. We combine geometric cues from 3D part fields, appearance cues from multi-view vision features, and semantic knowledge from language-model-generated affordance descriptions. Text-alignment loss ensures partlets share embedding space with text, enabling a theoretically open-vocabulary matching setup, given sufficient data. Our efficient and novel, one-shot, 3D part segmentation and naming method finds applications in several downstream tasks, including serving as a scalable annotation engine. As our model supports zero-shot matching to arbitrary descriptions and confidence-calibrated predictions for known categories, with human verification, we create a unified ontology that aligns PartNet, 3DCoMPaT++, and Find3D, consisting of 1,794 unique 3D parts. We introduce two novel metrics appropriate for the named 3D part segmentation task. We also show examples from our newly created TexParts dataset. --- ### The Subject of Emergent Misalignment in Superintelligence: An Anthropological, Cognitive Neuropsychological, Machine-Learning, and Ontological Perspective Authors: Muhammad Osama Imran, Roshni Lulla, Rodney Sappington Published: 2025-12-19 Categories: q-bio.NC, cs.AI arXiv: https://arxiv.org/abs/2512.17989v2 We examine the conceptual and ethical gaps in current representations of Superintelligence misalignment. We find throughout Superintelligence discourse an absent human subject, and an under-developed theorization of an "AI unconscious" that together are potentiality laying the groundwork for anti-social harm. With the rise of AI Safety that has both thematic potential for establishing pro-social and anti-social potential outcomes, we ask: what place does the human subject occupy in these imaginaries? How is human subjecthood positioned within narratives of catastrophic failure or rapid "takeoff" toward superintelligence? On another register, we ask: what unconscious or repressed dimensions are being inscribed into large-scale AI models? Are we to blame these agents in opting for deceptive strategies when undesirable patterns are inherent within our beings? In tracing these psychic and epistemic absences, our project calls for re-centering the human subject as the unstable ground upon which the ethical, unconscious, and misaligned dimensions of both human and machinic intelligence are co-constituted. Emergent misalignment cannot be understood solely through technical diagnostics typical of contemporary machine-learning safety research. Instead, it represents a multi-layered crisis. The human subject disappears not only through computational abstraction but through sociotechnical imaginaries that prioritize scalability, acceleration, and efficiency over vulnerability, finitude, and relationality. Likewise, the AI unconscious emerges not as a metaphor but as a structural reality of modern deep learning systems: vast latent spaces, opaque pattern formation, recursive symbolic play, and evaluation-sensitive behavior that surpasses explicit programming. These dynamics necessitate a reframing of misalignment as a relational instability embedded within human-machine ecologies. --- ### Enhancing Tea Leaf Disease Recognition with Attention Mechanisms and Grad-CAM Visualization Authors: Omar Faruq Shikdar, Fahad Ahammed, B. M. Shahria Alam, Golam Kibria, Tawhidur Rahman, Nishat Tasnim Niloy Published: 2025-12-19 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.17987v1 Tea is among the most widely consumed drinks globally. Tea production is a key industry for many countries. One of the main challenges in tea harvesting is tea leaf diseases. If the spread of tea leaf diseases is not stopped in time, it can lead to massive economic losses for farmers. Therefore, it is crucial to identify tea leaf diseases as soon as possible. Manually identifying tea leaf disease is an ineffective and time-consuming method, without any guarantee of success. Automating this process will improve both the efficiency and the success rate of identifying tea leaf diseases. The purpose of this study is to create an automated system that can classify different kinds of tea leaf diseases, allowing farmers to take action to minimize the damage. A novel dataset was developed specifically for this study. The dataset contains 5278 images across seven classes. The dataset was pre-processed prior to training the model. We deployed three pretrained models: DenseNet, Inception, and EfficientNet. EfficientNet was used only in the ensemble model. We utilized two different attention modules to improve model performance. The ensemble model achieved the highest accuracy of 85.68%. Explainable AI was introduced for better model interpretability. --- ### FedOAED: Federated On-Device Autoencoder Denoiser for Heterogeneous Data under Limited Client Availability Authors: S M Ruhul Kabir Howlader, Xiao Chen, Yifei Xie, Lu Liu Published: 2025-12-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.17986v1 Over the last few decades, machine learning (ML) and deep learning (DL) solutions have demonstrated their potential across many applications by leveraging large amounts of high-quality data. However, strict data-sharing regulations such as the General Data Protection Regulation (GDPR) and the Health Insurance Portability and Accountability Act (HIPAA) have prevented many data-driven applications from being realised. Federated Learning (FL), in which raw data never leaves local devices, has shown promise in overcoming these limitations. Although FL has grown rapidly in recent years, it still struggles with heterogeneity, which produces gradient noise, client-drift, and increased variance from partial client participation. In this paper, we propose FedOAED, a novel federated learning algorithm designed to mitigate client-drift arising from multiple local training updates and the variance induced by partial client participation. FedOAED incorporates an on-device autoencoder denoiser on the client side to mitigate client-drift and variance resulting from heterogeneous data under limited client availability. Experiments on multiple vision datasets under Non-IID settings demonstrate that FedOAED consistently outperforms state-of-the-art baselines. --- ### MoE-TransMov: A Transformer-based Model for Next POI Prediction in Familiar & Unfamiliar Movements Authors: Ruichen Tan, Jiawei Xue, Kota Tsubouchi, Takahiro Yabe, Satish V. Ukkusuri Published: 2025-12-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.17985v1 Accurate prediction of the next point of interest (POI) within human mobility trajectories is essential for location-based services, as it enables more timely and personalized recommendations. In particular, with the rise of these approaches, studies have shown that users exhibit different POI choices in their familiar and unfamiliar areas, highlighting the importance of incorporating user familiarity into predictive models. However, existing methods often fail to distinguish between the movements of users in familiar and unfamiliar regions. To address this, we propose MoE-TransMov, a Transformer-based model with a Transformer model with a Mixture-of-Experts (MoE) architecture designed to use one framework to capture distinct mobility patterns across different moving contexts without requiring separate training for certain data. Using user-check-in data, we classify movements into familiar and unfamiliar categories and develop a specialized expert network to improve prediction accuracy. Our approach integrates self-attention mechanisms and adaptive gating networks to dynamically select the most relevant expert models for different mobility contexts. Experiments on two real-world datasets, including the widely used but small open-source Foursquare NYC dataset and the large-scale Kyoto dataset collected with LY Corporation (Yahoo Japan Corporation), show that MoE-TransMov outperforms state-of-the-art baselines with notable improvements in Top-1, Top-5, Top-10 accuracy, and mean reciprocal rank (MRR). Given the results, we find that by using this approach, we can efficiently improve mobility predictions under different moving contexts, thereby enhancing the personalization of recommendation systems and advancing various urban applications. --- ### A Hybrid Inductive-Transductive Network for Traffic Flow Imputation on Unsampled Locations Authors: Mohammadmahdi Rahimiasl, Ynte Vanderhoydonc, Siegfried Mercelis Published: 2025-12-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.17984v1 Accurately imputing traffic flow at unsensed locations is difficult: loop detectors provide precise but sparse measurements, speed from probe vehicles is widely available yet only weakly correlated with flow, and nearby links often exhibit strong heterophily in the scale of traffic flow (e.g., ramps vs. mainline), which breaks standard GNN assumptions. We propose HINT, a Hybrid INductive-Transductive Network, and an INDU-TRANSDUCTIVE training strategy that treats speed as a transductive, network-wide signal while learning flow inductively to generalize to unseen locations. HINT couples (i) an inductive spatial transformer that learns similarity-driven, long-range interactions from node features with (ii) a diffusion GCN conditioned by FiLM on rich static context (OSM-derived attributes and traffic simulation), and (iii) a node-wise calibration layer that corrects scale biases per segment. Training uses masked reconstruction with epoch-wise node sampling, hard-node mining to emphasize difficult sensors, and noise injection on visible flows to prevent identity mapping, while graph structure is built from driving distances. Across three real-world datasets, MOW (Antwerp, Belgium), UTD19-Torino, and UTD19-Essen, HINT consistently surpasses state-of-the-art inductive baselines. Relative to KITS, HINT reduces MAE on MOW by $\approx42$% with basic simulation and $\approx50$% with calibrated simulation; on Torino by $\approx22$%, and on Essen by $\approx12$%. Even without simulation, HINT remains superior on MOW and Torino, while simulation is crucial on Essen. These results show that combining inductive flow imputation with transductive speed, traffic simulations and external geospatial improves accuracy for the task described above. --- ### Parameter-Efficient Fine-Tuning for HAR: Integrating LoRA and QLoRA into Transformer Models Authors: Irina Seregina, Philippe Lalanda, German Vega Published: 2025-12-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.17983v1 Human Activity Recognition is a foundational task in pervasive computing. While recent advances in self-supervised learning and transformer-based architectures have significantly improved HAR performance, adapting large pretrained models to new domains remains a practical challenge due to limited computational resources on target devices. This papers investigates parameter-efficient fine-tuning techniques, specifically Low-Rank Adaptation (LoRA) and Quantized LoRA, as scalable alternatives to full model fine-tuning for HAR. We propose an adaptation framework built upon a Masked Autoencoder backbone and evaluate its performance under a Leave-One-Dataset-Out validation protocol across five open HAR datasets. Our experiments demonstrate that both LoRA and QLoRA can match the recognition performance of full fine-tuning while significantly reducing the number of trainable parameters, memory usage, and training time. Further analyses reveal that LoRA maintains robust performance even under limited supervision and that the adapter rank provides a controllable trade-off between accuracy and efficiency. QLoRA extends these benefits by reducing the memory footprint of frozen weights through quantization, with minimal impact on classification quality. --- ### Adaptive Agents in Spatial Double-Auction Markets: Modeling the Emergence of Industrial Symbiosis Authors: Matthieu Mastio, Paul Saves, Benoit Gaudou, Nicolas Verstaevel Published: 2025-12-19 Categories: cs.GT, cs.AI, cs.MA, econ.GN, stat.AP arXiv: https://arxiv.org/abs/2512.17979v2 Industrial symbiosis fosters circularity by enabling firms to repurpose residual resources, yet its emergence is constrained by socio-spatial frictions that shape costs, matching opportunities, and market efficiency. Existing models often overlook the interaction between spatial structure, market design, and adaptive firm behavior, limiting our understanding of where and how symbiosis arises. We develop an agent-based model where heterogeneous firms trade byproducts through a spatially embedded double-auction market, with prices and quantities emerging endogenously from local interactions. Leveraging reinforcement learning, firms adapt their bidding strategies to maximize profit while accounting for transport costs, disposal penalties, and resource scarcity. Simulation experiments reveal the economic and spatial conditions under which decentralized exchanges converge toward stable and efficient outcomes. Counterfactual regret analysis shows that sellers' strategies approach a near Nash equilibrium, while sensitivity analysis highlights how spatial structures and market parameters jointly govern circularity. Our model provides a basis for exploring policy interventions that seek to align firm incentives with sustainability goals, and more broadly demonstrates how decentralized coordination can emerge from adaptive agents in spatially constrained markets. --- ### MEGState: Phoneme Decoding from Magnetoencephalography Signals Authors: Shuntaro Suzuki, Chia-Chun Dan Hsu, Yu Tsao, Komei Sugiura Published: 2025-12-19 Categories: q-bio.NC, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2512.17978v1 Decoding linguistically meaningful representations from non-invasive neural recordings remains a central challenge in neural speech decoding. Among available neuroimaging modalities, magnetoencephalography (MEG) provides a safe and repeatable means of mapping speech-related cortical dynamics, yet its low signal-to-noise ratio and high temporal dimensionality continue to hinder robust decoding. In this work, we introduce MEGState, a novel architecture for phoneme decoding from MEG signals that captures fine-grained cortical responses evoked by auditory stimuli. Extensive experiments on the LibriBrain dataset demonstrate that MEGState consistently surpasses baseline model across multiple evaluation metrics. These findings highlight the potential of MEG-based phoneme decoding as a scalable pathway toward non-invasive brain-computer interfaces for speech. --- ### Sampling from multimodal distributions with warm starts: Non-asymptotic bounds for the Reweighted Annealed Leap-Point Sampler Authors: Holden Lee, Matheau Santana-Gijzen Published: 2025-12-19 Categories: stat.ML, cs.LG, math.PR, math.ST, stat.CO arXiv: https://arxiv.org/abs/2512.17977v2 Sampling from multimodal distributions is a central challenge in Bayesian inference and machine learning. In light of hardness results for sampling -- classical MCMC methods, even with tempering, can suffer from exponential mixing times -- a natural question is how to leverage additional information, such as a warm start point for each mode, to enable faster mixing across modes. To address this, we introduce Reweighted ALPS (Re-ALPS), a modified version of the Annealed Leap-Point Sampler (ALPS) that dispenses with the Gaussian approximation assumption. We prove the first polynomial-time bound that works in a general setting, under a natural assumption that each component contains significant mass relative to the others when tilted towards the corresponding warm start point. Similarly to ALPS, we define distributions tilted towards a mixture centered at the warm start points, and at the coldest level, use teleportation between warm start points to enable efficient mixing across modes. In contrast to ALPS, our method does not require Hessian information at the modes, but instead estimates component partition functions via Monte Carlo. This additional estimation step is crucial in allowing the algorithm to handle target distributions with more complex geometries besides approximate Gaussian. For the proof, we show convergence results for Markov processes when only part of the stationary distribution is well-mixing and estimation for partition functions for individual components of a mixture. --- ### Re-assessing the evidence for mental rotation abilities in children using computational models Authors: Arthur Aubret, Jochen Triesch Published: 2025-12-19 Categories: q-bio.NC, cs.AI arXiv: https://arxiv.org/abs/2512.17972v1 There is strong and diverse evidence for mental rotation (MR) abilities in adults. However, current evidence for MR in children rests on just a few behavioral paradigms adapted from the adult literature. Here, we leverage recent computational models of the development of children's object recognition abilities to re-assess the evidence for MR in children. The computational models simulate infants' acquisition of object representations during embodied interactions with objects. We consider two different object recognition strategies, different from MRs, and assess their ability to replicate results from three classical MR tasks assigned to children between the ages of 6 months and 5 years. Our results show that MR may play no role in producing the results obtained from children younger than 5 years. In fact, we find that a simple recognition strategy that reflects a pixel-wise comparison of stimuli is sufficient to model children's behavior in the most used MR task. Thus, our study reopens the debate on how and when children develop genuine MR abilities. --- ### CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs Authors: Gunho Park, Jeongin Bae, Byeongwook Kim, Baeseong park, Jiwon Ryu, Hoseung Kim, Se Jung Kwon, Dongsoo Lee Published: 2025-12-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.17970v1 Weight-only quantization is widely used to mitigate the memory-bound nature of LLM inference. Codebook-based methods extend this trend by achieving strong accuracy in the extremely low-bit regime (e.g., 2-bit). However, current kernels rely on dequantization, which repeatedly fetches centroids and reconstructs weights, incurring substantial latency and cache pressure. We present CodeGEMM, a codebook-centric GEMM kernel that replaces dequantization with precomputed inner products between centroids and activations stored in a lightweight Psumbook. At inference, code indices directly gather these partial sums, eliminating per-element lookups and reducing the on-chip footprint. The kernel supports the systematic exploration of latency-memory-accuracy trade-offs under a unified implementation. On Llama-3 models, CodeGEMM delivers 1.83x (8B) and 8.93x (70B) speedups in the 2-bit configuration compared to state-of-the-art codebook-based quantization at comparable accuracy and further improves computing efficiency and memory subsystem utilization. --- ### Convolutional-neural-operator-based transfer learning for solving PDEs Authors: Peng Fan, Guofei Pang Published: 2025-12-19 Categories: cs.LG, cs.AI, math-ph arXiv: https://arxiv.org/abs/2512.17969v1 Convolutional neural operator is a CNN-based architecture recently proposed to enforce structure-preserving continuous-discrete equivalence and enable the genuine, alias-free learning of solution operators of PDEs. This neural operator was demonstrated to outperform for certain cases some baseline models such as DeepONet, Fourier neural operator, and Galerkin transformer in terms of surrogate accuracy. The convolutional neural operator, however, seems not to be validated for few-shot learning. We extend the model to few-shot learning scenarios by first pre-training a convolutional neural operator using a source dataset and then adjusting the parameters of the trained neural operator using only a small target dataset. We investigate three strategies for adjusting the parameters of a trained neural operator, including fine-tuning, low-rank adaption, and neuron linear transformation, and find that the neuron linear transformation strategy enjoys the highest surrogate accuracy in solving PDEs such as Kuramoto-Sivashinsky equation, Brusselator diffusion-reaction system, and Navier-Stokes equations. ---