# Research Papers — 2025-11 · bundle page 172 of 197 > Full-text markdown of papers #8551–8600 of 9830 published in 2025-11, newest first. AI-readable mirrors of the full 9830-paper 2025-11 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2025-11/page/172 [← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/171.md) [Next bundle →](https://aipod.dev/papers/archive/2025-11/page/173.md) ### Depth-induced NTK: Bridging Over-parameterized Neural Networks and Deep Neural Kernels Authors: Yong-Ming Tian, Shuang Liang, Shao-Qun Zhang, Feng-Lei Fan Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.05585v1 While deep learning has achieved remarkable success across a wide range of applications, its theoretical understanding of representation learning remains limited. Deep neural kernels provide a principled framework to interpret over-parameterized neural networks by mapping hierarchical feature transformations into kernel spaces, thereby combining the expressive power of deep architectures with the analytical tractability of kernel methods. Recent advances, particularly neural tangent kernels (NTKs) derived by gradient inner products, have established connections between infinitely wide neural networks and nonparametric Bayesian inference. However, the existing NTK paradigm has been predominantly confined to the infinite-width regime, while overlooking the representational role of network depth. To address this gap, we propose a depth-induced NTK kernel based on a shortcut-related architecture, which converges to a Gaussian process as the network depth approaches infinity. We theoretically analyze the training invariance and spectrum properties of the proposed kernel, which stabilizes the kernel dynamics and mitigates degeneration. Experimental results further underscore the effectiveness of our proposed method. Our findings significantly extend the existing landscape of the neural kernel theory and provide an in-depth understanding of deep learning and the scaling law. --- ### Beyond Resolution: Multi-Scale Weather and Climate Data for Alpine Renewable Energy in the Digital Twin Era -- First Evaluations and Recommendations Authors: Irene Schicker, Marianne Bügelmayer-Blaschek, Annemarie Lexer, Katharina Baier, Kristofer Hasel, Paolo Gazzaneo Published: 2025-11-05 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2511.05584v1 When Austrian hydropower production plummeted by 44% in early 2025 due to reduced snowpack, it exposed a critical vulnerability: standard meteorological and climatological datasets systematically fail in mountain regions that hold untapped renewable potential. This perspectives paper evaluates emerging solutions to the Alpine energy-climate data gap, analyzing datasets from global reanalyses (ERA5, 31 km) to kilometre-scale Digital Twins (Climate DT, Extremes DT, 4.4 km), regional reanalyses (ARA, 2.5 km), and next-generation AI weather prediction models (AIFS, 31 km). The multi-resolution assessment reveals that no single dataset excels universally: coarse reanalyses provide essential climatologies but miss valley-scale processes, while Digital Twins resolve Alpine dynamics yet remain computationally demanding. Effective energy planning therefore requires strategic dataset combinations validated against energy-relevant indices such as population-weighted extremes, wind-gust return periods, and Alpine-adjusted storm thresholds. A key frontier is sub-hourly (10-15 min) temporal resolution to match grid-operation needs. Six evidence-based recommendations outline pathways for bridging spatial and temporal scales. As renewable deployment expands globally into complex terrain, the Alpine region offers transferable perspectives for tackling identical forecasting and climate analysis challenges in mountainous regions worldwide. --- ### Uncertainty Modeling for Multi-Objective RTA Interception with Distillation Acceleration Authors: Gaoxiang Zhao, Ruinan Qiu, Xiaoting Wang, Pengpeng Zhao, Rongjin Wang, Xiaoting Wang, Zhangang Lin, Xiaoqiang Wang Published: 2025-11-05 Categories: cs.LG, cs.GT arXiv: https://arxiv.org/abs/2511.05582v3 Real-Time Auction (RTA) interception decides which incoming advertising requests reach downstream systems, and therefore controls the quality of the data those systems learn from. At JD.com, off-site advertising produces on the order of hundreds of billions of requests per day, and the RTA channel alone serves up to hundreds of millions of requests per minute. Filtering low-quality and fraudulent traffic at this scale requires estimating each request's value with calibrated confidence, which we treat as an uncertainty modeling problem. Two obstacles stand in the way. First, advertising labels are severely imbalanced: deals are rare, and we show both analytically and empirically that standard weight-based uncertainty degrades under such sparsity, collapsing onto predicted probability and adding no signal. Second, methods such as SWAG and Bayesian neural networks require multiple stochastic forward passes per request, making full-traffic scoring prohibitively expensive. We address both problems with UMDA, a multi-objective framework that shares uncertainty across funnel-correlated objectives, using the reliable uncertainty of a balanced metric to compensate for the degenerate uncertainty of sparse ones. We then distill the multi-pass teacher into a single-pass student that reproduces both aleatoric and epistemic uncertainty at roughly one-tenth of the inference cost. On JD e-commerce dataset and the public Criteo dataset, UMDA supplies more effective samples to downstream tasks, and the distilled student preserves this capability. In production, it scores the full traffic in a near-line pipeline that feeds an hourly blacklist for online interception; a seven-day A/B test on 5% of live traffic cuts the click fraud rate by 3.59% and raises CVR by 4.01% at a matched interception ratio while leaving converted users essentially unchanged, and the model has since been deployed to full traffic. --- ### Cannistraci-Hebb Training on Ultra-Sparse Spiking Neural Networks Authors: Yuan Hua, Jilin Zhang, Yingtao Zhang, Wenqi Gu, Leyi You, Baobo Xiong, Carlo Vittorio Cannistraci, Hong Chen Published: 2025-11-05 Categories: cs.NE arXiv: https://arxiv.org/abs/2511.05581v1 Inspired by the brain's spike-based computation, spiking neural networks (SNNs) inherently possess temporal activation sparsity. However, when it comes to the sparse training of SNNs in the structural connection domain, existing methods fail to achieve ultra-sparse network structures without significant performance loss, thereby hindering progress in energy-efficient neuromorphic computing. This limitation presents a critical challenge: how to achieve high levels of structural connection sparsity while maintaining performance comparable to fully connected networks. To address this challenge, we propose the Cannistraci-Hebb Spiking Neural Network (CH-SNN), a novel and generalizable dynamic sparse training framework for SNNs consisting of four stages. First, we propose a sparse spike correlated topological initialization (SSCTI) method to initialize a sparse network based on node correlations. Second, temporal activation sparsity and structural connection sparsity are integrated via a proposed sparse spike weight initialization (SSWI) method. Third, a hybrid link removal score (LRS) is applied to prune redundant weights and inactive neurons, improving information flow. Finally, the CH3-L3 network automaton framework inspired by Cannistraci-Hebb learning theory is incorporated to perform link prediction for potential synaptic regrowth. These mechanisms enable CH-SNN to achieve sparsification across all linear layers. We have conducted extensive experiments on six datasets including CIFAR-10 and CIFAR-100, evaluating various network architectures such as spiking convolutional neural networks and Spikformer. --- ### Approximating the Mathematical Structure of Psychodynamics Authors: Bryce-Allen Bagley, Navin Khoshnan Published: 2025-11-05 Categories: q-bio.NC, cs.CL, cs.CY, cs.HC arXiv: https://arxiv.org/abs/2511.05580v1 The complexity of human cognition has meant that psychology makes more use of theory and conceptual models than perhaps any other biomedical field. To enable precise quantitative study of the full breadth of phenomena in psychological and psychiatric medicine as well as cognitive aspects of AI safety, there is a need for a mathematical formulation which is both mathematically precise and equally accessible to experts from numerous fields. In this paper we formalize human psychodynamics via the diagrammatic framework of process theory, describe its key properties, and explain the links between a diagrammatic representation and central concepts in analysis of cognitive processes in contexts such as psychotherapy, neurotechnology, AI alignment, AI agent representation of individuals in autonomous negotiations, developing human-like AI systems, and other aspects of AI safety. --- ### UTF-8 Plumbing: Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8 Authors: Preston Firestone, Shubham Ugare, Gagandeep Singh, Sasa Misailovic Published: 2025-11-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.05578v1 Subword tokenization segments input text according to a pre-defined vocabulary to feed it into a language model; the language model, in turn, generates a sequence made from this same vocabulary. The members of the vocabulary can be built of code points or bytes. Using code points means that all members of the vocabulary are valid UTF-8 characters. However, it also requires thousands of initial members to achieve acceptable coverage of inputs. Beginning with bytes, on the contrary, avoids out-of-vocabulary errors with only 256 initial members of the vocabulary, but the members of the vocabulary and sequences of them are not guaranteed to be valid UTF-8. Sequences that are not valid UTF-8 break code that assumes its input to be valid UTF-8. Applications of language models must account for the breakage thereby introduced. In this paper, we formalize tokenization using monoid theory and prove that tokenizers whose vocabularies contain tokens that are ill-formed UTF-8 can always produce sequences that are ill-formed UTF-8. We demonstrate formally that attempting to incrementally convert tokens back to a string and interpret the results as UTF-8 gives different results than converting the whole sequence of tokens at once. This formal result predicts real-world bugs: we evaluate mitigations for the problem identified and provide case studies of major foundation models, serving engines, and constrained generation systems. --- ### SWAP: Towards Copyright Auditing of Soft Prompts via Sequential Watermarking Authors: Wenyuan Yang, Yichen Sun, Changzheng Chen, Zhixuan Chu, Jiaheng Zhang, Yiming Li, Dacheng Tao Published: 2025-11-05 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.04711v2 Large-scale vision-language models, especially CLIP, have demonstrated remarkable performance across diverse downstream tasks. Soft prompts, as carefully crafted modules that efficiently adapt vision-language models to specific tasks, necessitate effective copyright protection. In this paper, we investigate model copyright protection by auditing whether suspicious third-party models incorporate protected soft prompts. While this can be viewed as a special case of model ownership auditing, our analysis shows that existing techniques are ineffective due to prompt learning's unique characteristics. Non-intrusive auditing is inherently prone to false positives when independent models share similar data distributions with victim models. Intrusive approaches also fail: backdoor methods designed for CLIP cannot embed functional triggers, while extending traditional DNN backdoor techniques to prompt learning suffers from harmfulness and ambiguity challenges. We find that these failures in intrusive auditing stem from the same fundamental reason: watermarking operates within the same decision space as the primary task yet pursues opposing objectives. Motivated by these findings, we propose sequential watermarking for soft prompts (SWAP), which implants watermarks into a different and more complex space. SWAP encodes watermarks through a specific order of defender-specified out-of-distribution classes, inspired by the zero-shot prediction capability of CLIP. This watermark, which is embedded in a more complex space, keeps the original prediction label unchanged, making it less opposed to the primary task. We further design a hypothesis-test-guided verification protocol for SWAP and provide a theoretical analysis of when verification works. Extensive experiments on 11 datasets demonstrate SWAP's effectiveness, harmlessness, and robustness against potential attacks. --- ### GEMMA-SQL: A Novel Text-to-SQL Model Based on Large Language Models Authors: Hari Mohan Pandey, Anshul Gupta, Subham Sarkar, Minakshi Tomer, Schneider Johannes, Yan Gong Published: 2025-11-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.04710v1 Text-to-SQL systems enable users to interact with structured databases using natural language, eliminating the need for specialized programming knowledge. In this work, we introduce GEMMA-SQL, a lightweight and efficient text-to-SQL model built upon the open-source Gemma 2B architecture. Unlike many large language models (LLMs), GEMMA-SQL is fine-tuned in a resource-efficient, iterative manner and can be deployed on low-cost hardware. Leveraging the SPIDER benchmark for training and evaluation, GEMMA-SQL combines multiple prompting strategies, including few-shot learning, to enhance SQL query generation accuracy. The instruction-tuned variant, GEMMA-SQL Instruct, achieves 66.8% Test-Suite accuracy and 63.3% Exact Set Match accuracy, outperforming several state-of-the-art baselines such as IRNet, RYANSQL, and CodeXDavinci. The proposed approach demonstrates that effective prompt design and targeted instruction tuning can significantly boost performance while maintaining high scalability and adaptability. These results position GEMMA-SQL as a practical, open-source alternative for robust and accessible text-to-SQL systems. --- ### Jailbreaking in the Haystack Authors: Rishi Rajesh Shah, Chen Henry Wu, Shashwat Saxena, Ziqian Zhong, Alexander Robey, Aditi Raghunathan Published: 2025-11-05 Categories: cs.CR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2511.04707v2 Recent advances in long-context language models (LMs) have enabled million-token inputs, expanding their capabilities across complex tasks like computer-use agents. Yet, the safety implications of these extended contexts remain unclear. To bridge this gap, we introduce NINJA (short for Needle-in-haystack jailbreak attack), a method that jailbreaks aligned LMs by appending benign, model-generated content to harmful user goals. Critical to our method is the observation that the position of harmful goals play an important role in safety. Experiments on standard safety benchmark, HarmBench, show that NINJA significantly increases attack success rates across state-of-the-art open and proprietary models, including LLaMA, Qwen, Mistral, and Gemini. Unlike prior jailbreaking methods, our approach is low-resource, transferable, and less detectable. Moreover, we show that NINJA is compute-optimal -- under a fixed compute budget, increasing context length can outperform increasing the number of trials in best-of-N jailbreak. These findings reveal that even benign long contexts -- when crafted with careful goal positioning -- introduce fundamental vulnerabilities in modern LMs. --- ### The Human Flourishing Geographic Index: A County-Level Dataset for the United States, 2013--2023 Authors: Stefano M. Iacus, Devika Jain, Andrea Nasuto, Giuseppe Porro, Marcello Carammia, Andrea Vezzulli Published: 2025-11-05 Categories: cs.CL, cs.CY, stat.AP arXiv: https://arxiv.org/abs/2511.03915v1 Quantifying human flourishing, a multidimensional construct including happiness, health, purpose, virtue, relationships, and financial stability, is critical for understanding societal well-being beyond economic indicators. Existing measures often lack fine spatial and temporal resolution. Here we introduce the Human Flourishing Geographic Index (HFGI), derived from analyzing approximately 2.6 billion geolocated U.S. tweets (2013-2023) using fine-tuned large language models to classify expressions across 48 indicators aligned with Harvard's Global Flourishing Study framework plus attitudes towards migration and perception of corruption. The dataset offers monthly and yearly county- and state-level indicators of flourishing-related discourse, validated to confirm that the measures accurately represent the underlying constructs and show expected correlations with established indicators. This resource enables multidisciplinary analyses of well-being, inequality, and social change at unprecedented resolution, offering insights into the dynamics of human flourishing as reflected in social media discourse across the United States over the past decade. --- ### Inference-Time Optimization of Prompt Embeddings in Diffusion Models: A Comparison of sep-CMA-ES and Adam Authors: Domício Pereira Neto, João Correia, Penousal Machado Published: 2025-11-05 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2511.03913v3 Deep diffusion models have revolutionized image generation by producing high-quality outputs. However, achieving specific objectives with these models often requires costly adaptations such as fine-tuning, which can be resource-intensive and time-consuming. An alternative approach is inference-time control, which involves optimizing the prompt embeddings to guide the generation process without altering the model weights. We explore prompt-embedding search optimization for the Stable Diffusion XL Turbo model, comparing a gradient-free evolutionary approach, the Separable Covariance Matrix Adaptation Evolution Strategy (sep-CMA-ES), against the widely used gradient-based optimizer Adaptive Moment Estimation (Adam). Candidate images are evaluated by a weighted objective that combines LAION Aesthetic Predictor V2 and CLIPScore, enabling explicit trade-offs between aesthetic quality and prompt-image alignment. On 36 prompts sampled from Parti Prompts (P2) under three weight settings (aesthetics-only, balanced, alignment-only), sep-CMA-ES consistently achieves higher objective values than Adam. We additionally analyze divergence from the unoptimized baseline using cosine similarity and SSIM and report the compute and memory footprints. These results suggest that sep-CMA-ES is an effective inference-time optimizer for prompt-embedding search, improving aesthetics-alignment trade-offs and resource usage without model fine-tuning. --- ### I Detect What I Don't Know: Incremental Anomaly Learning with Stochastic Weight Averaging-Gaussian for Oracle-Free Medical Imaging Authors: Nand Kumar Yadav, Rodrigue Rizk, William CW Chen, KC Santosh Published: 2025-11-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.03912v2 Unknown anomaly detection in medical imaging remains a fundamental challenge due to the scarcity of labeled anomalies and the high cost of expert supervision. We introduce an unsupervised, oracle-free framework that incrementally expands a trusted set of normal samples without any anomaly labels. Starting from a small, verified seed of normal images, our method alternates between lightweight adapter updates and uncertainty-gated sample admission. A frozen pretrained vision backbone is augmented with tiny convolutional adapters, ensuring rapid domain adaptation with negligible computational overhead. Extracted embeddings are stored in a compact coreset enabling efficient k-nearest neighbor anomaly (k-NN) scoring. Safety during incremental expansion is enforced by dual probabilistic gates, a sample is admitted into the normal memory only if its distance to the existing coreset lies within a calibrated z-score threshold, and its SWAG-based epistemic uncertainty remains below a seed-calibrated bound. This mechanism prevents drift and false inclusions without relying on generative reconstruction or replay buffers. Empirically, our system steadily refines the notion of normality as unlabeled data arrive, producing substantial gains over baselines. On COVID-CXR, ROC-AUC improves from 0.9489 to 0.9982 (F1: 0.8048 to 0.9746); on Pneumonia CXR, ROC-AUC rises from 0.6834 to 0.8968; and on Brain MRI ND-5, ROC-AUC increases from 0.6041 to 0.7269 and PR-AUC from 0.7539 to 0.8211. These results highlight the effectiveness and efficiency of the proposed framework for real-world, label-scarce medical imaging applications. --- ### DecoHD: Decomposed Hyperdimensional Classification under Extreme Memory Budgets Authors: Sanggeon Yun, Hyunwoo Oh, Ryozo Masukawa, Mohsen Imani Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03911v2 Decomposition is a proven way to shrink deep networks without changing input-output dimensionality or interface semantics. We bring this idea to hyperdimensional computing (HDC), where footprint cuts usually shrink the feature axis and erode concentration and robustness. Prior HDC decompositions decode via fixed atomic hypervectors, which are ill-suited for compressing learned class prototypes. We introduce DecoHD, which learns directly in a decomposed HDC parameterization: a small, shared set of per-layer channels with multiplicative binding across layers and bundling at the end, yielding a large representational space from compact factors. DecoHD compresses along the class axis via a lightweight bundling head while preserving native bind-bundle-score; training is end-to-end, and inference remains pure HDC, aligning with in/near-memory accelerators. In evaluation, DecoHD attains extreme memory savings with only minor accuracy degradation under tight deployment budgets. On average it stays within about 0.1-0.15% of a strong non-reduced HDC baseline (worst case 5.7%), is more robust to random bit-flip noise, reaches its accuracy plateau with up to ~97% fewer trainable parameters, and--in hardware--delivers roughly 277x/35x energy/speed gains over a CPU (AMD Ryzen 9 9950X), 13.5x/3.7x over a GPU (NVIDIA RTX 4090), and 2.0x/2.4x over a baseline HDC ASIC. --- ### Tensor Computation of Euler Characteristic Functions and Transforms Authors: Jessi Cisewski-Kehe, Brittany Terese Fasy, Alexander McCleary, Eli Quist Published: 2025-11-05 Categories: cs.CG, cs.LG, math.AT arXiv: https://arxiv.org/abs/2511.03909v3 The weighted Euler characteristic transform (WECT) and Euler characteristic function (ECF) have proven to be useful tools in a variety of applications. However, current methods for computing these functions are either not optimized for GPU computation or do not scale to higher-dimensional settings. In this work, we present a tensor-based framework for computing such topological descriptors which is highly optimized for GPU architectures and works in full generality across simplicial and cubical complexes of arbitrary dimension. Experimentally, the framework demonstrates significant speedups over existing methods when computing the WECT and ECF across a variety of two- and three-dimensional datasets. Computation of these transforms is implemented in a publicly available Python package called pyECT. --- ### Context informs pragmatic interpretation in vision-language models Authors: Alvin Wei Ming Tan, Ben Prystawski, Veronica Boyce, Michael C. Frank Published: 2025-11-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.03908v1 Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, order, and relevance. Without relevant context, models were above chance but substantially worse than humans. However, with relevant context, model performance increased dramatically over trials. Few-shot reference games with abstract referents remain a difficult task for machine learning models. --- ### SnappyMeal: Design and Longitudinal Evaluation of a Multimodal AI Food Logging Application Authors: Liam Bakar, Zachary Englhardt, Vidya Srinivas, Girish Narayanswamy, Dilini Nissanka, Shwetak Patel, Vikram Iyer Published: 2025-11-05 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2511.03907v1 Food logging, both self-directed and prescribed, plays a critical role in uncovering correlations between diet, medical, fitness, and health outcomes. Through conversations with nutritional experts and individuals who practice dietary tracking, we find current logging methods, such as handwritten and app-based journaling, are inflexible and result in low adherence and potentially inaccurate nutritional summaries. These findings, corroborated by prior literature, emphasize the urgent need for improved food logging methods. In response, we propose SnappyMeal, an AI-powered dietary tracking system that leverages multimodal inputs to enable users to more flexibly log their food intake. SnappyMeal introduces goal-dependent follow-up questions to intelligently seek missing context from the user and information retrieval from user grocery receipts and nutritional databases to improve accuracy. We evaluate SnappyMeal through publicly available nutrition benchmarks and a multi-user, 3-week, in-the-wild deployment capturing over 500 logged food instances. Users strongly praised the multiple available input methods and reported a strong perceived accuracy. These insights suggest that multimodal AI systems can be leveraged to significantly improve dietary tracking flexibility and context-awareness, laying the groundwork for a new class of intelligent self-tracking applications. --- ### GRAD: Graph-Retrieved Adaptive Decoding for Hallucination Mitigation Authors: Manh Nguyen, Sunil Gupta, Dai Do, Hung Le Published: 2025-11-05 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2511.03900v1 Hallucination mitigation remains a persistent challenge for large language models (LLMs), even as model scales grow. Existing approaches often rely on external knowledge sources, such as structured databases or knowledge graphs, accessed through prompting or retrieval. However, prompt-based grounding is fragile and domain-sensitive, while symbolic knowledge integration incurs heavy retrieval and formatting costs. Motivated by knowledge graphs, we introduce Graph-Retrieved Adaptive Decoding (GRAD), a decoding-time method that grounds generation in corpus-derived evidence without retraining. GRAD constructs a sparse token transition graph by accumulating next-token logits across a small retrieved corpus in a single forward pass. During decoding, graph-retrieved logits are max-normalized and adaptively fused with model logits to favor high-evidence continuations while preserving fluency. Across three models and a range of question-answering benchmarks spanning intrinsic, extrinsic hallucination, and factuality tasks, GRAD consistently surpasses baselines, achieving up to 9.7$\%$ higher intrinsic accuracy, 8.6$\%$ lower hallucination rates, and 6.9$\%$ greater correctness compared to greedy decoding, while attaining the highest truth--informativeness product score among all methods. GRAD offers a lightweight, plug-and-play alternative to contrastive decoding and knowledge graph augmentation, demonstrating that statistical evidence from corpus-level token transitions can effectively steer generation toward more truthful and verifiable outputs. --- ### Secure Code Generation at Scale with Reflexion Authors: Arup Datta, Ahmed Aljohani, Hyunsook Do Published: 2025-11-05 Categories: cs.CR, cs.AI, cs.CE, cs.SE arXiv: https://arxiv.org/abs/2511.03898v1 Large language models (LLMs) are now widely used to draft and refactor code, but code that works is not necessarily secure. We evaluate secure code generation using the Instruct Prime, which eliminated compliance-required prompts and cue contamination, and evaluate five instruction-tuned code LLMs using a zero-shot baseline and a three-round reflexion prompting approach. Security is measured using the Insecure Code Detector (ICD), and results are reported by measuring Repair, Regression, and NetGain metrics, considering the programming language and CWE family. Our findings show that insecurity remains common at the first round: roughly 25-33% of programs are insecure at a zero-shot baseline (t0 ). Weak cryptography/config-dependent bugs are the hardest to avoid while templated ones like XSS, code injection, and hard-coded secrets are handled more reliably. Python yields the highest secure rates; C and C# are the lowest, with Java, JS, PHP, and C++ in the middle. Reflexion prompting improves security for all models, improving average accuracy from 70.74% at t0 to 79.43% at t3 , with the largest gains in the first round followed by diminishing returns. The trends with Repair, Regression, and NetGain metrics show that applying one to two rounds produces most of the benefits. A replication package is available at https://doi.org/10.5281/zenodo.17065846. --- ### A general technique for approximating high-dimensional empirical kernel matrices Authors: Chiraag Kaushik, Justin Romberg, Vidya Muthukumar Published: 2025-11-05 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2511.03892v1 We present simple, user-friendly bounds for the expected operator norm of a random kernel matrix under general conditions on the kernel function $k(\cdot,\cdot)$. Our approach uses decoupling results for U-statistics and the non-commutative Khintchine inequality to obtain upper and lower bounds depending only on scalar statistics of the kernel function and a ``correlation kernel'' matrix corresponding to $k(\cdot,\cdot)$. We then apply our method to provide new, tighter approximations for inner-product kernel matrices on general high-dimensional data, where the sample size and data dimension are polynomially related. Our method obtains simplified proofs of existing results that rely on the moment method and combinatorial arguments while also providing novel approximation results for the case of anisotropic Gaussian data. Finally, using similar techniques to our approximation result, we show a tighter lower bound on the bias of kernel regression with anisotropic Gaussian data. --- ### Improving Diagnostic Performance on Small and Imbalanced Datasets Using Class-Based Input Image Composition Authors: Hlali Azzeddine, Majid Ben Yakhlef, Soulaiman El Hazzat Published: 2025-11-05 Categories: cs.CV, cs.AI, cs.DB arXiv: https://arxiv.org/abs/2511.03891v2 Small, imbalanced datasets and poor input image quality can lead to high false predictions rates with deep learning models. This paper introduces Class-Based Image Composition, an approach that allows us to reformulate training inputs through a fusion of multiple images of the same class into combined visual composites, named Composite Input Images (CoImg). That enhances the intra-class variance and improves the valuable information density per training sample and increases the ability of the model to distinguish between subtle disease patterns. Our method was evaluated on the Optical Coherence Tomography Dataset for Image-Based Deep Learning Methods (OCTDL) (Kulyabin et al., 2024), which contains 2,064 high-resolution optical coherence tomography (OCT) scans of the human retina, representing seven distinct diseases with a significant class imbalance. We constructed a perfectly class-balanced version of this dataset, named Co-OCTDL, where each scan is resented as a 3x1 layout composite image. To assess the effectiveness of this new representation, we conducted a comparative analysis between the original dataset and its variant using a VGG16 model. A fair comparison was ensured by utilizing the identical model architecture and hyperparameters for all experiments. The proposed approach markedly improved diagnostic results.The enhanced Dataset achieved near-perfect accuracy (99.6%) with F1-score (0.995) and AUC (0.9996), compared to a baseline model trained on raw dataset. The false prediction rate was also significantly lower, this demonstrates that the method can producehigh-quality predictions even for weak datasets affected by class imbalance or small sample size. --- ### Shape Deformation Networks for Automated Aortic Valve Finite Element Meshing from 3D CT Images Authors: Linchen Qian, Jiasong Chen, Ruonan Gong, Wei Sun, Minliang Liu, Liang Liang Published: 2025-11-05 Categories: eess.IV, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.03890v1 Accurate geometric modeling of the aortic valve from 3D CT images is essential for biomechanical analysis and patient-specific simulations to assess valve health or make a preoperative plan. However, it remains challenging to generate aortic valve meshes with both high-quality and consistency across different patients. Traditional approaches often produce triangular meshes with irregular topologies, which can result in poorly shaped elements and inconsistent correspondence due to inter-patient anatomical variation. In this work, we address these challenges by introducing a template-fitting pipeline with deep neural networks to generate structured quad (i.e., quadrilateral) meshes from 3D CT images to represent aortic valve geometries. By remeshing aortic valves of all patients with a common quad mesh template, we ensure a uniform mesh topology with consistent node-to-node and element-to-element correspondence across patients. This consistency enables us to simplify the learning objective of the deep neural networks, by employing a loss function with only two terms (i.e., a geometry reconstruction term and a smoothness regularization term), which is sufficient to preserve mesh smoothness and element quality. Our experiments demonstrate that the proposed approach produces high-quality aortic valve surface meshes with improved smoothness and shape quality, while requiring fewer explicit regularization terms compared to the traditional methods. These results highlight that using structured quad meshes for the template and neural network training not only ensures mesh correspondence and quality but also simplifies the training process, thus enhancing the effectiveness and efficiency of aortic valve modeling. --- ### YOLO-SAT: A Data-based and Model-based Enhanced YOLOv12 Model for Desert Waste Detection and Classification Authors: Abdulmumin Sa'ad, Sulaimon Oyeniyi Adebayo Published: 2025-11-05 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.03888v2 The global waste crisis is escalating, with solid waste generation expected to increase tremendously in the coming years. Traditional waste collection methods, particularly in remote or harsh environments like deserts, are labor-intensive, inefficient, and often hazardous. Recent advances in computer vision and deep learning have opened the door to automated waste detection systems, yet most research focuses on urban environments and recyclable materials, overlooking organic and hazardous waste and underexplored terrains such as deserts. In this work, we propose YOLO-SAT, an enhanced real-time object detection framework based on a pruned, lightweight version of YOLOv12 integrated with Self-Adversarial Training (SAT) and specialized data augmentation strategies. Using the DroneTrashNet dataset, we demonstrate significant improvements in precision, recall, and mean average precision (mAP), while achieving low latency and compact model size suitable for deployment on resource-constrained aerial drones. Benchmarking YOLO-SAT against state-of-the-art lightweight YOLO variants further highlights its optimal balance of accuracy and efficiency. Our results validate the effectiveness of combining data-centric and model-centric enhancements for robust, real-time waste detection in desert environments. --- ### Investigating Robot Control Policy Learning for Autonomous X-ray-guided Spine Procedures Authors: Florence Klitzner, Blanca Inigo, Benjamin D. Killeen, Lalithkumar Seenivasan, Michelle Song, Axel Krieger, Mathias Unberath Published: 2025-11-05 Categories: cs.CV, cs.AI, cs.LG, cs.RO arXiv: https://arxiv.org/abs/2511.03882v2 Imitation learning-based robot control policies are enjoying renewed interest in video-based robotics. However, it remains unclear whether this approach applies to X-ray-guided procedures, such as spine instrumentation, with sparse inputs. We examine the feasibility, opportunities and challenges for imitation policy learning in bi-plane-guided cannula insertion. We develop an in silico sandbox for scalable, automated simulation of X-ray-guided spine procedures with a high degree of realism. We curate a dataset of correct trajectories and corresponding bi-planar X-ray sequences that emulate the stepwise alignment of providers. We then train imitation learning policies for planning and open-loop control that iteratively align a cannula in a vertebroplasty setting solely based on visual information. This precisely controlled setup offers insights into limitations and capabilities of this method. Our policy succeeded on the first attempt in 68.5% of cases, maintaining safe intra-pedicular trajectories across diverse vertebral levels. The policy transferred to complex anatomy, including fractures, as well as varied anatomies and initializations. Rollouts on real X-ray indicate that partial sim-to-real transfer with plausible trajectories is possible. While these preliminary results are promising, we also identify limitations, especially in entry point precision. The current results present a clear benchmark for future efforts, while with more robust priors and domain knowledge, such models may provide a foundation for future efforts toward lightweight and CT-free robotic intra-operative spinal navigation. --- ### Evaluating Machine Translation Datasets for Low-Web Data Languages: A Gendered Lens Authors: Hellina Hailu Nigatu, Bethelhem Yemane Mamo, Bontu Fufa Balcha, Debora Taye Tesfaye, Elbethel Daniel Zewdie, Ikram Behiru Nesiru, Jitu Ewnetu Hailu, Senait Mengesha Yayo Published: 2025-11-05 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2511.03880v1 As low-resourced languages are increasingly incorporated into NLP research, there is an emphasis on collecting large-scale datasets. But in prioritizing quantity over quality, we risk 1) building language technologies that perform poorly for these languages and 2) producing harmful content that perpetuates societal biases. In this paper, we investigate the quality of Machine Translation (MT) datasets for three low-resourced languages--Afan Oromo, Amharic, and Tigrinya, with a focus on the gender representation in the datasets. Our findings demonstrate that while training data has a large representation of political and religious domain text, benchmark datasets are focused on news, health, and sports. We also found a large skew towards the male gender--in names of persons, the grammatical gender of verbs, and in stereotypical depictions in the datasets. Further, we found harmful and toxic depictions against women, which were more prominent for the language with the largest amount of data, underscoring that quantity does not guarantee quality. We hope that our work inspires further inquiry into the datasets collected for low-resourced languages and prompts early mitigation of harmful content. WARNING: This paper contains discussion of NSFW content that some may find disturbing. --- ### KnowThyself: An Agentic Assistant for LLM Interpretability Authors: Suraj Prasai, Mengnan Du, Ying Zhang, Fan Yang Published: 2025-11-05 Categories: cs.AI, cs.IR, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2511.03878v1 We develop KnowThyself, an agentic assistant that advances large language model (LLM) interpretability. Existing tools provide useful insights but remain fragmented and code-intensive. KnowThyself consolidates these capabilities into a chat-based interface, where users can upload models, pose natural language questions, and obtain interactive visualizations with guided explanations. At its core, an orchestrator LLM first reformulates user queries, an agent router further directs them to specialized modules, and the outputs are finally contextualized into coherent explanations. This design lowers technical barriers and provides an extensible platform for LLM inspection. By embedding the whole process into a conversational workflow, KnowThyself offers a robust foundation for accessible LLM interpretability. --- ### Benchmark Datasets for Lead-Lag Forecasting on Social Platforms Authors: Kimia Kazemian, Zhenzhen Liu, Yangfanyu Yang, Katie Luo, Shuhan Gu, Audrey Du, Xinyu Yang, Jack Jansons, Kilian Q. Weinberger, John Thickstun, Yian Yin, Sarah Dean Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03877v2 Social and collaborative platforms emit multivariate time-series traces in which early interactions -- such as views, likes, or downloads -- are followed, sometimes months or years later, by higher impact like citations, sales, or reviews. We formalize this setting as Lead-Lag Forecasting (LLF): given an early usage channel (the lead), predict a correlated but temporally shifted outcome channel (the lag). Despite the ubiquity of such patterns, LLF has not been treated as a unified forecasting problem within the time-series community, largely due to the absence of standardised datasets. To anchor research in LLF, here we present two high-volume benchmark datasets: arXiv (accesses -> citations of 2.3M papers) and GitHub (pushes/stars -> forks of 3M repositories). Our datasets provide ideal testbeds for lead-lag forecasting, by capturing long-horizon dynamics across years, spanning the full spectrum of outcomes, and avoiding survivorship bias in sampling. We documented all technical details of data curation and cleaning, verified the presence of lead-lag dynamics through statistical and classification tests, and benchmarked parametric and non-parametric baselines for regression. Our study establishes LLF as a novel forecasting paradigm and lays an empirical foundation for its systematic exploration in social and usage data. --- ### Computed Tomography (CT)-derived Cardiovascular Flow Estimation Using Physics-Informed Neural Networks Improves with Sinogram-based Training: A Simulation Study Authors: Jinyuxuan Guo, Gurnoor Singh Khurana, Alejandro Gonzalo Grande, Juan C. del Alamo, Francisco Contijoch Published: 2025-11-05 Categories: eess.IV, cs.CV, cs.LG, physics.med-ph arXiv: https://arxiv.org/abs/2511.03876v1 Background: Non-invasive imaging-based assessment of blood flow plays a critical role in evaluating heart function and structure. Computed Tomography (CT) is a widely-used imaging modality that can robustly evaluate cardiovascular anatomy and function, but direct methods to estimate blood flow velocity from movies of contrast evolution have not been developed. Purpose: This study evaluates the impact of CT imaging on Physics-Informed Neural Networks (PINN)-based flow estimation and proposes an improved framework, SinoFlow, which uses sinogram data directly to estimate blood flow. Methods: We generated pulsatile flow fields in an idealized 2D vessel bifurcation using computational fluid dynamics and simulated CT scans with varying gantry rotation speeds, tube currents, and pulse mode imaging settings. We compared the performance of PINN-based flow estimation using reconstructed images (ImageFlow) to SinoFlow. Results: SinoFlow significantly improved flow estimation performance by avoiding propagating errors introduced by filtered backprojection. SinoFlow was robust across all tested gantry rotation speeds and consistently produced lower mean squared error and velocity errors than ImageFlow. Additionally, SinoFlow was compatible with pulsed-mode imaging and maintained higher accuracy with shorter pulse widths. Conclusions: This study demonstrates the potential of SinoFlow for CT-based flow estimation, providing a more promising approach for non-invasive blood flow assessment. The findings aim to inform future applications of PINNs to CT images and provide a solution for image-based estimation, with reasonable acquisition parameters yielding accurate flow estimates. --- ### OMPILOT: Harnessing Transformer Models for Auto Parallelization to Shared Memory Computing Paradigms Authors: Arijit Bhattacharjee, Ali TehraniJamsaz, Le Chen, Niranjan Hasabnis, Mihai Capota, Nesreen Ahmed, Ali Jannesari Published: 2025-11-05 Categories: cs.DC, cs.AI, cs.LG, cs.PF, cs.PL arXiv: https://arxiv.org/abs/2511.03866v2 Recent advances in large language models (LLMs) have significantly accelerated progress in code translation, enabling more accurate and efficient transformation across programming languages. While originally developed for natural language processing, LLMs have shown strong capabilities in modeling programming language syntax and semantics, outperforming traditional rule-based systems in both accuracy and flexibility. These models have streamlined cross-language conversion, reduced development overhead, and accelerated legacy code migration. In this paper, we introduce OMPILOT, a novel domain-specific encoder-decoder transformer tailored for translating C++ code into OpenMP, enabling effective shared-memory parallelization. OMPILOT leverages custom pre-training objectives that incorporate the semantics of parallel constructs and combines both unsupervised and supervised learning strategies to improve code translation robustness. Unlike previous work that focused primarily on loop-level transformations, OMPILOT operates at the function level to capture a wider semantic context. To evaluate our approach, we propose OMPBLEU, a novel composite metric specifically crafted to assess the correctness and quality of OpenMP parallel constructs, addressing limitations in conventional translation metrics. --- ### Levers of Power in the Field of AI Authors: Tammy Mackenzie, Sukriti Punj, Natalie Perez, Sreyoshi Bhaduri, Branislav Radeljic Published: 2025-11-05 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2511.03859v1 This paper examines how decision makers in academia, government, business, and civil society navigate questions of power in implementations of artificial intelligence. The study explores how individuals experience and exercise levers of power, which are presented as social mechanisms that shape institutional responses to technological change. The study reports on the responses of personalized questionnaires designed to gather insight on a decision maker's institutional purview, based on an institutional governance framework developed from the work of Neo-institutionalists. Findings present the anonymized, real responses and circumstances of respondents in the form of twelve fictional personas of high-level decision makers from North America and Europe. These personas illustrate how personal agency, organizational logics, and institutional infrastructures may intersect in the governance of AI. The decision makers' responses to the questionnaires then inform a discussion of the field-level personal power of decision makers, methods of fostering institutional stability in times of change, and methods of influencing institutional change in the field of AI. The final section of the discussion presents a table of the dynamics of the levers of power in the field of AI for change makers and five testable hypotheses for institutional and social movement researchers. In summary, this study provides insight on the means for policymakers within institutions and their counterparts in civil society to personally engage with AI governance. --- ### Noise Injection: Improving Out-of-Distribution Generalization for Limited Size Datasets Authors: Duong Mai, Lawrence Hall Published: 2025-11-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.03855v2 Deep learned (DL) models for image recognition have been shown to fail to generalize to data from different devices, populations, etc. COVID-19 detection from Chest X-rays (CXRs), in particular, has been shown to fail to generalize to out-of-distribution (OOD) data from new clinical sources not covered in the training set. This occurs because models learn to exploit shortcuts - source-specific artifacts that do not translate to new distributions - rather than reasonable biomarkers to maximize performance on in-distribution (ID) data. Rendering the models more robust to distribution shifts, our study investigates the use of fundamental noise injection techniques (Gaussian, Speckle, Poisson, and Salt and Pepper) during training. Our empirical results demonstrate that this technique can significantly reduce the performance gap between ID and OOD evaluation from 0.10-0.20 to 0.01-0.06, based on results averaged over ten random seeds across key metrics such as AUC, F1, accuracy, recall and specificity. Our source code is publicly available at https://github.com/Duongmai127/Noisy-ood --- ### Which Similarity-Sensitive Entropy (Sentropy)? Authors: Phuc Nguyen, Josiah Couch, Rahul Bansal, Alexandra Morgan, Chris Tam, Miao Li, Rima Arnaout, Ramy Arnaout Published: 2025-11-05 Categories: cs.IT, cs.LG, q-bio.PE arXiv: https://arxiv.org/abs/2511.03849v4 Shannon entropy is not the only entropy that is relevant to machine-learning datasets, nor possibly even the most important one. Traditional entropies such as Shannon entropy capture information represented by elements' frequencies but not the richer information encoded by their similarities and differences. Capturing the latter requires similarity-sensitive entropy (``sentropy''). Sentropy can be measured using either the recently developed Leinster-Cobbold-Reeve framework (LCR) or the newer Vendi score (VS). This raises the practical question of which one to use: LCR or VS. Here we address this question theoretically and numerically, using 53 large and well-known imaging and tabular datasets. We find that LCR and VS values can differ by orders of magnitude and are complementary, except in limiting cases. We show that both LCR and VS results depend on how similarities are scaled, and introduce the notion of ``half-distance'' to parameterize this dependence. We prove the VS provides an upper bound on LCR for all non-negative values of the Rényi-Hill order parameter, as well as for negative values in the special case that the similarity matrix is full rank. We conclude that VS is preferable only when a dataset's elements can be usefully interpreted as linear combinations of a more fundamental set of ``ur-elements'' or when the system that the dataset describes has a quantum-mechanical character. In the broader case where one simply wishes to capture the rich information encoded by elements' similarities and differences as well as their frequencies, we propose that LCR should be favored; nevertheless, for certain half-distances the two methods can complement each other. --- ### To See or To Read: User Behavior Reasoning in Multimodal LLMs Authors: Tianning Dong, Luyi Ma, Varun Vasudevan, Jason Cho, Sushant Kumar, Kannan Achan Published: 2025-11-05 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.03845v1 Multimodal Large Language Models (MLLMs) are reshaping how modern agentic systems reason over sequential user-behavior data. However, whether textual or image representations of user behavior data are more effective for maximizing MLLM performance remains underexplored. We present \texttt{BehaviorLens}, a systematic benchmarking framework for assessing modality trade-offs in user-behavior reasoning across six MLLMs by representing transaction data as (1) a text paragraph, (2) a scatter plot, and (3) a flowchart. Using a real-world purchase-sequence dataset, we find that when data is represented as images, MLLMs next-purchase prediction accuracy is improved by 87.5% compared with an equivalent textual representation without any additional computational cost. --- ### Enhancing Q-Value Updates in Deep Q-Learning via Successor-State Prediction Authors: Lipeng Zu, Hansong Zhou, Xiaonan Zhang Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03836v2 Deep Q-Networks (DQNs) estimate future returns by learning from transitions sampled from a replay buffer. However, the target updates in DQN often rely on next states generated by actions from past, potentially suboptimal, policy. As a result, these states may not provide informative learning signals, causing high variance into the update process. This issue is exacerbated when the sampled transitions are poorly aligned with the agent's current policy. To address this limitation, we propose the Successor-state Aggregation Deep Q-Network (SADQ), which explicitly models environment dynamics using a stochastic transition model. SADQ integrates successor-state distributions into the Q-value estimation process, enabling more stable and policy-aligned value updates. Additionally, it explores a more efficient action selection strategy with the modeled transition structure. We provide theoretical guarantees that SADQ maintains unbiased value estimates while reducing training variance. Our extensive empirical results across standard RL benchmarks and real-world vector-based control tasks demonstrate that SADQ consistently outperforms DQN variants in both stability and learning efficiency. --- ### Higher-Order Causal Structure Learning with Additive Models Authors: James Enouen, Yujia Zheng, Ignavier Ng, Yan Liu, Kun Zhang Published: 2025-11-05 Categories: cs.LG, math.ST, stat.ML arXiv: https://arxiv.org/abs/2511.03831v1 Causal structure learning has long been the central task of inferring causal insights from data. Despite the abundance of real-world processes exhibiting higher-order mechanisms, however, an explicit treatment of interactions in causal discovery has received little attention. In this work, we focus on extending the causal additive model (CAM) to additive models with higher-order interactions. This second level of modularity we introduce to the structure learning problem is most easily represented by a directed acyclic hypergraph which extends the DAG. We introduce the necessary definitions and theoretical tools to handle the novel structure we introduce and then provide identifiability results for the hyper DAG, extending the typical Markov equivalence classes. We next provide insights into why learning the more complex hypergraph structure may actually lead to better empirical results. In particular, more restrictive assumptions like CAM correspond to easier-to-learn hyper DAGs and better finite sample complexity. We finally develop an extension of the greedy CAM algorithm which can handle the more complex hyper DAG search space and demonstrate its empirical usefulness in synthetic experiments. --- ### Divide, Cache, Conquer: Dichotomic Prompting for Efficient Multi-Label LLM-Based Classification Authors: Mikołaj Langner, Jan Eliasz, Ewa Rudnicka, Jan Kocoń Published: 2025-11-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.03830v1 We introduce a method for efficient multi-label text classification with large language models (LLMs), built on reformulating classification tasks as sequences of dichotomic (yes/no) decisions. Instead of generating all labels in a single structured response, each target dimension is queried independently, which, combined with a prefix caching mechanism, yields substantial efficiency gains for short-text inference without loss of accuracy. To demonstrate the approach, we focus on affective text analysis, covering 24 dimensions including emotions and sentiment. Using LLM-to-SLM distillation, a powerful annotator model (DeepSeek-V3) provides multiple annotations per text, which are aggregated to fine-tune smaller models (HerBERT-Large, CLARIN-1B, PLLuM-8B, Gemma3-1B). The fine-tuned models show significant improvements over zero-shot baselines, particularly on the dimensions seen during training. Our findings suggest that decomposing multi-label classification into dichotomic queries, combined with distillation and cache-aware inference, offers a scalable and effective framework for LLM-based classification. While we validate the method on affective states, the approach is general and applicable across domains. --- ### From Static Constraints to Dynamic Adaptation: Sample-Level Constraint Relaxation for Offline-to-Online Reinforcement Learning Authors: Lipeng Zu, Yu Qian, Shayok Chakraborty, Xiaonan Zhang Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03828v3 Offline-to-online reinforcement learning (O2O RL) faces a central challenge between retaining offline conservatism and adapting to online feedback under distribution shift. This challenge arises because data behavior evolves during fine-tuning, rendering data origin a misleading basis for constraint handling and thereby leading to objective-data mismatch. We therefore propose Dynamic Alignment for RElaxation (DARE), a distribution-aware framework for sample-level constraint relaxation based on the behavioral consistency with a behavior model. To our knowledge, DARE is the first to condition constraint relaxation on behavioral consistency via a posterior-induced exchange mechanism, moving beyond a binary offline/online data distinction. Importantly, DARE requires only per-sample behavioral alignment, enabling instantiation on top of many offline algorithms with flexible choices of behavior models and fine-tuning objectives. We provide a theoretical analysis showing that behavior-based sample exchange consistently improves the distinction between offline-like and online-like subsets. Experiments on D4RL demonstrate that DARE consistently improves fine-tuning stability and achieves superior final performance over strong offline-to-online baselines. (The code is publicly available at \url{https://github.com/lpzu/DARE}.) --- ### STARS: Synchronous Token Alignment for Robust Supervision in Large Language Models Authors: Mohammad Atif Quamar, Mohammad Areeb, Mikhail Kuznetsov, Muslum Ozgur Ozmen, Z. Berkay Celik Published: 2025-11-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.03827v2 Aligning large language models (LLMs) with human values is crucial for safe deployment. Inference-time techniques offer granular control over generation; however, they rely on model uncertainty, meaning an internal estimate of how likely the model believes its next tokens or outputs are correct, for segmentation. We show that this introduces two critical limitations: (a) vulnerability to miscalibrated confident hallucinations and (b) poor hardware utilization due to asynchronous, ragged batch processing. Together, these issues reduce alignment reliability while increasing token and compute costs, which limits their practical scalability. To address these limitations, building on dynamic inference-time alignment methods, we introduce STARS, Synchronous Token Alignment for Robust Supervision, a decoding-time algorithm, which steers generation by enforcing verification at fixed-horizon intervals. By decoupling segmentation from confidence, STARS enables lockstep parallel execution and robustly detects errors that uncertainty metrics miss. On the HH-RLHF benchmark, we demonstrate that STARS achieves competitive alignment quality with that of state-of-the-art dynamic methods, while strictly bounding rejection costs and maximizing system throughput. Furthermore, it outperforms fine-tuning and several state-of-the-art inference-time decoding strategies by good margins, and establishes fixed-horizon sampling as a robust, system-efficient alternative for aligning LLMs at scale. The code is publicly available at https://github.com/purseclab/STARS. --- ### CORE -- A Cell-Level Coarse-to-Fine Image Registration Engine for Multi-stain Image Alignment Authors: Esha Sadia Nasir, Behnaz Elhaminia, Mark Eastwood, Catherine King, Owen Cain, Lorraine Harper, Paul Moss, Dimitrios Chanouzas, David Snead, Nasir Rajpoot, Adam Shephard, Shan E Ahmed Raza Published: 2025-11-05 Categories: q-bio.QM, cs.AI arXiv: https://arxiv.org/abs/2511.03826v4 Accurate and efficient registration of whole slide images (WSIs) is essential for high-resolution, nuclei-level analysis in multi-stained tissue slides. We propose a novel coarse-to-fine framework CORE for accurate nuclei-level registration across diverse multimodal whole-slide image (WSI) datasets. The coarse registration stage leverages prompt-based tissue mask extraction to effectively filter out artefacts and non-tissue regions, followed by global alignment using tissue morphology and accelerated dense feature matching with a pre-trained feature extractor. From the coarsely aligned slides, nuclei centroids are detected and subjected to fine-grained rigid registration using a custom, shape-aware point-set registration model. Finally, non-rigid alignment at the cellular level is achieved by estimating a non-linear displacement field using Coherent Point Drift (CPD). Our approach benefits from automatically generated nuclei that enhance the accuracy of deformable registration and ensure precise nuclei-level correspondence across modalities. The proposed model is evaluated on three publicly available WSI registration datasets, and two private datasets. We show that CORE outperforms current state-of-the-art methods in terms of generalisability, precision, and robustness in bright-field and immunofluorescence microscopy WSIs --- ### How Different Tokenization Algorithms Impact LLMs and Transformer Models for Binary Code Analysis Authors: Ahmed Mostafa, Raisul Arefin Nahid, Samuel Mulder Published: 2025-11-05 Categories: cs.AI, cs.CL, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2511.03825v1 Tokenization is fundamental in assembly code analysis, impacting intrinsic characteristics like vocabulary size, semantic coverage, and extrinsic performance in downstream tasks. Despite its significance, tokenization in the context of assembly code remains an underexplored area. This study aims to address this gap by evaluating the intrinsic properties of Natural Language Processing (NLP) tokenization models and parameter choices, such as vocabulary size. We explore preprocessing customization options and pre-tokenization rules tailored to the unique characteristics of assembly code. Additionally, we assess their impact on downstream tasks like function signature prediction -- a critical problem in binary code analysis. To this end, we conduct a thorough study on various tokenization models, systematically analyzing their efficiency in encoding assembly instructions and capturing semantic nuances. Through intrinsic evaluations, we compare tokenizers based on tokenization efficiency, vocabulary compression, and representational fidelity for assembly code. Using state-of-the-art pre-trained models such as the decoder-only Large Language Model (LLM) Llama 3.2, the encoder-only transformer BERT, and the encoder-decoder model BART, we evaluate the effectiveness of these tokenizers across multiple performance metrics. Preliminary findings indicate that tokenizer choice significantly influences downstream performance, with intrinsic metrics providing partial but incomplete predictability of extrinsic evaluation outcomes. These results reveal complex trade-offs between intrinsic tokenizer properties and their utility in practical assembly code tasks. Ultimately, this study provides valuable insights into optimizing tokenization models for low-level code analysis, contributing to the robustness and scalability of Natural Language Model (NLM)-based binary analysis workflows. --- ### Sketch-Augmented Features Improve Learning Long-Range Dependencies in Graph Neural Networks Authors: Ryien Hosseini, Filippo Simini, Venkatram Vishwanath, Rebecca Willett, Henry Hoffmann Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03824v1 Graph Neural Networks learn on graph-structured data by iteratively aggregating local neighborhood information. While this local message passing paradigm imparts a powerful inductive bias and exploits graph sparsity, it also yields three key challenges: (i) oversquashing of long-range information, (ii) oversmoothing of node representations, and (iii) limited expressive power. In this work we inject randomized global embeddings of node features, which we term \textit{Sketched Random Features}, into standard GNNs, enabling them to efficiently capture long-range dependencies. The embeddings are unique, distance-sensitive, and topology-agnostic -- properties which we analytically and empirically show alleviate the aforementioned limitations when injected into GNNs. Experimental results on real-world graph learning tasks confirm that this strategy consistently improves performance over baseline GNNs, offering both a standalone solution and a complementary enhancement to existing techniques such as graph positional encodings. Our source code is available at \href{https://github.com/ryienh/sketched-random-features}{https://github.com/ryienh/sketched-random-features}. --- ### PLLuM: A Family of Polish Large Language Models Authors: Jan Kocoń, Maciej Piasecki, Arkadiusz Janz, Teddy Ferdinan, Łukasz Radliński, Bartłomiej Koptyra, Marcin Oleksy, Stanisław Woźniak, Paweł Walkowiak, Konrad Wojtasik, Julia Moska, Tomasz Naskręt et al. Published: 2025-11-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2511.03823v1 Large Language Models (LLMs) play a central role in modern artificial intelligence, yet their development has been primarily focused on English, resulting in limited support for other languages. We present PLLuM (Polish Large Language Model), the largest open-source family of foundation models tailored specifically for the Polish language. Developed by a consortium of major Polish research institutions, PLLuM addresses the need for high-quality, transparent, and culturally relevant language models beyond the English-centric commercial landscape. We describe the development process, including the construction of a new 140-billion-token Polish text corpus for pre-training, a 77k custom instructions dataset, and a 100k preference optimization dataset. A key component is a Responsible AI framework that incorporates strict data governance and a hybrid module for output correction and safety filtering. We detail the models' architecture, training procedures, and alignment techniques for both base and instruction-tuned variants, and demonstrate their utility in a downstream task within public administration. By releasing these models publicly, PLLuM aims to foster open research and strengthen sovereign AI technologies in Poland. --- ### SiLVi: Simple Interface for Labeling Video Interactions Authors: Ozan Kanbertay, Richard Vogg, Elif Karakoc, Peter M. Kappeler, Claudia Fichtel, Alexander S. Ecker Published: 2025-11-05 Categories: cs.CV, q-bio.QM arXiv: https://arxiv.org/abs/2511.03819v2 Computer vision methods are increasingly used for the automated analysis of large volumes of video data collected through camera traps, drones, or direct observations of animals in the wild. While recent advances have focused primarily on detecting individual actions, much less work has addressed the detection and annotation of interactions -- a crucial aspect for understanding social and individualized animal behavior. Existing open-source annotation tools support either behavioral labeling without localization of individuals, or localization without the capacity to capture interactions. To bridge this gap, we present SiLVi, an open-source labeling software that integrates both functionalities. SiLVi enables researchers to annotate behaviors and interactions directly within video data, generating structured outputs suitable for training and validating computer vision models. By linking behavioral ecology with computer vision, SiLVi facilitates the development of automated approaches for fine-grained behavioral analyses. Although developed primarily in the context of animal behavior, SiLVi could be useful more broadly to annotate human interactions in other videos that require extracting dynamic scene graphs. The software, along with documentation and download instructions, is available at: https://silvi.eckerlab.org. --- ### One Size Does Not Fit All: Architecture-Aware Adaptive Batch Scheduling with DEBA Authors: François Belias, Naser Ezzati-Jivan, Foutse Khomh Published: 2025-11-05 Categories: cs.LG, cs.PF arXiv: https://arxiv.org/abs/2511.03809v1 Adaptive batch size methods aim to accelerate neural network training, but existing approaches apply identical adaptation strategies across all architectures, assuming a one-size-fits-all solution. We introduce DEBA (Dynamic Efficient Batch Adaptation), an adaptive batch scheduler that monitors gradient variance, gradient norm variation and loss variation to guide batch size adaptations. Through systematic evaluation across six architectures (ResNet-18/50, DenseNet-121, EfficientNet-B0, MobileNet-V3, ViT-B16) on CIFAR-10 and CIFAR-100, with five random seeds per configuration, we demonstrate that the architecture fundamentally determines adaptation efficacy. Our findings reveal that: (1) lightweight and medium-depth architectures (MobileNet-V3, DenseNet-121, EfficientNet-B0) achieve a 45-62% training speedup with simultaneous accuracy improvements of 1-7%; (2) shallow residual networks (ResNet-18) show consistent gains of +2.4 - 4.0% in accuracy, 36 - 43% in speedup, while deep residual networks (ResNet-50) exhibit high variance and occasional degradation; (3) already-stable architectures (ViT-B16) show minimal speedup (6%) despite maintaining accuracy, indicating that adaptation benefits vary with baseline optimization characteristics. We introduce a baseline characterization framework using gradient stability metrics (stability score, gradient norm variation) that predicts which architectures will benefit from adaptive scheduling. Our ablation studies reveal critical design choices often overlooked in prior work: sliding window statistics (vs. full history) and sufficient cooldown periods (5+ epochs) between adaptations are essential for success. This work challenges the prevailing assumption that adaptive methods generalize across architectures and provides the first systematic evidence that batch size adaptation requires an architecture-aware design. --- ### Optimizing Reasoning Efficiency through Prompt Difficulty Prediction Authors: Bo Zhao, Berkcan Kapusuzoglu, Kartik Balasubramaniam, Sambit Sahu, Supriyo Chakraborty, Genta Indra Winata Published: 2025-11-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.03808v1 Reasoning language models perform well on complex tasks but are costly to deploy due to their size and long reasoning traces. We propose a routing approach that assigns each problem to the smallest model likely to solve it, reducing compute without sacrificing accuracy. Using intermediate representations from s1.1-32B, we train lightweight predictors of problem difficulty or model correctness to guide routing across a pool of reasoning models. On diverse math benchmarks, routing improves efficiency over random assignment and matches s1.1-32B's performance while using significantly less compute. Our results demonstrate that difficulty-aware routing is effective for cost-efficient deployment of reasoning models. --- ### Fair and Explainable Credit-Scoring under Concept Drift: Adaptive Explanation Frameworks for Evolving Populations Authors: Shivogo John Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03807v1 Evolving borrower behaviors, shifting economic conditions, and changing regulatory landscapes continuously reshape the data distributions underlying modern credit-scoring systems. Conventional explainability techniques, such as SHAP, assume static data and fixed background distributions, making their explanations unstable and potentially unfair when concept drift occurs. This study addresses that challenge by developing adaptive explanation frameworks that recalibrate interpretability and fairness in dynamically evolving credit models. Using a multi-year credit dataset, we integrate predictive modeling via XGBoost with three adaptive SHAP variants: (A) per-slice explanation reweighting that adjusts for feature distribution shifts, (B) drift-aware SHAP rebaselining with sliding-window background samples, and (C) online surrogate calibration using incremental Ridge regression. Each method is benchmarked against static SHAP explanations using metrics of predictive performance (AUC, F1), directional and rank stability (cosine, Kendall tau), and fairness (demographic parity and recalibration). Results show that adaptive methods, particularly rebaselined and surrogate-based explanations, substantially improve temporal stability and reduce disparate impact across demographic groups without degrading predictive accuracy. Robustness tests, including counterfactual perturbations, background sensitivity analysis, and proxy-variable detection, confirm the resilience of adaptive explanations under real-world drift conditions. These findings establish adaptive explainability as a practical mechanism for sustaining transparency, accountability, and ethical reliability in data-driven credit systems, and more broadly, in any domain where decision models evolve with population change. --- ### FusionDP: Foundation Model-Assisted Differentially Private Learning for Partially Sensitive Features Authors: Linghui Zeng, Ruixuan Liu, Atiquer Rahman Sarkar, Xiaoqian Jiang, Joyce C. Ho, Li Xiong Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03806v1 Ensuring the privacy of sensitive training data is crucial in privacy-preserving machine learning. However, in practical scenarios, privacy protection may be required for only a subset of features. For instance, in ICU data, demographic attributes like age and gender pose higher privacy risks due to their re-identification potential, whereas raw lab results are generally less sensitive. Traditional DP-SGD enforces privacy protection on all features in one sample, leading to excessive noise injection and significant utility degradation. We propose FusionDP, a two-step framework that enhances model utility under feature-level differential privacy. First, FusionDP leverages large foundation models to impute sensitive features given non-sensitive features, treating them as external priors that provide high-quality estimates of sensitive attributes without accessing the true values during model training. Second, we introduce a modified DP-SGD algorithm that trains models on both original and imputed features while formally preserving the privacy of the original sensitive features. We evaluate FusionDP on two modalities: a sepsis prediction task on tabular data from PhysioNet and a clinical note classification task from MIMIC-III. By comparing against privacy-preserving baselines, our results show that FusionDP significantly improves model performance while maintaining rigorous feature-level privacy, demonstrating the potential of foundation model-driven imputation to enhance the privacy-utility trade-off for various modalities. --- ### Learning Paths for Dynamic Measure Transport: A Control Perspective Authors: Aimee Maurais, Bamdad Hosseini, Youssef Marzouk Published: 2025-11-05 Categories: stat.ML, cs.LG, stat.CO arXiv: https://arxiv.org/abs/2511.03797v1 We bring a control perspective to the problem of identifying paths of measures for sampling via dynamic measure transport (DMT). We highlight the fact that commonly used paths may be poor choices for DMT and connect existing methods for learning alternate paths to mean-field games. Based on these connections we pose a flexible family of optimization problems for identifying tilted paths of measures for DMT and advocate for the use of objective terms which encourage smoothness of the corresponding velocities. We present a numerical algorithm for solving these problems based on recent Gaussian process methods for solution of partial differential equations and demonstrate the ability of our method to recover more efficient and smooth transport models compared to those which use an untilted reference path. --- ### Expert Evaluation of LLM World Models: A High-$T_c$ Superconductivity Case Study Authors: Haoyu Guo, Maria Tikhanovskaya, Paul Raccuglia, Alexey Vlaskin, Chris Co, Daniel J. Liebling, Scott Ellsworth, Matthew Abraham, Elizabeth Dorfman, N. P. Armitage, Chunhan Feng, Antoine Georges et al. Published: 2025-11-05 Categories: cond-mat.supr-con, cond-mat.str-el, cs.AI arXiv: https://arxiv.org/abs/2511.03782v1 Large Language Models (LLMs) show great promise as a powerful tool for scientific literature exploration. However, their effectiveness in providing scientifically accurate and comprehensive answers to complex questions within specialized domains remains an active area of research. Using the field of high-temperature cuprates as an exemplar, we evaluate the ability of LLM systems to understand the literature at the level of an expert. We construct an expert-curated database of 1,726 scientific papers that covers the history of the field, and a set of 67 expert-formulated questions that probe deep understanding of the literature. We then evaluate six different LLM-based systems for answering these questions, including both commercially available closed models and a custom retrieval-augmented generation (RAG) system capable of retrieving images alongside text. Experts then evaluate the answers of these systems against a rubric that assesses balanced perspectives, factual comprehensiveness, succinctness, and evidentiary support. Among the six systems two using RAG on curated literature outperformed existing closed models across key metrics, particularly in providing comprehensive and well-supported answers. We discuss promising aspects of LLM performances as well as critical short-comings of all the models. The set of expert-formulated questions and the rubric will be valuable for assessing expert level performance of LLM based reasoning systems. --- ### Contamination Detection for VLMs using Multi-Modal Semantic Perturbation Authors: Jaden Park, Mu Cai, Feng Yao, Jingbo Shang, Soochahn Lee, Yong Jae Lee Published: 2025-11-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.03774v2 Recent advances in Vision-Language Models (VLMs) have achieved state-of-the-art performance on numerous benchmark tasks. However, the use of internet-scale, often proprietary, pretraining corpora raises a critical concern for both practitioners and users: inflated performance due to test-set leakage. While prior works have proposed mitigation strategies such as decontamination of pretraining data and benchmark redesign for LLMs, the complementary direction of developing detection methods for contaminated VLMs remains underexplored. To address this gap, we deliberately contaminate open-source VLMs on popular benchmarks and show that existing detection approaches either fail outright or exhibit inconsistent behavior. We then propose a novel simple yet effective detection method based on multi-modal semantic perturbation, demonstrating that contaminated models fail to generalize under controlled perturbations. Finally, we validate our approach across multiple realistic contamination strategies, confirming its robustness and effectiveness. The code and perturbed dataset are released at https://github.com/jadenpark0/mm-perturb. --- ### Scaling Agent Learning via Experience Synthesis Authors: Zhaorun Chen, Zhuokai Zhao, Kai Zhang, Bo Liu, Qi Qi, Yifan Wu, Tarun Kalluri, Sara Cao, Yuanhao Xiong, Haibo Tong, Huaxiu Yao, Hengduo Li et al. Published: 2025-11-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2511.03773v2 While reinforcement learning (RL) can empower autonomous agents by enabling self-improvement through interaction, its practical adoption remains challenging due to costly rollouts, limited task diversity, unreliable reward signals, and infrastructure complexity, all of which obstruct the collection of scalable experience data. To address these challenges, we introduce DreamGym, the first unified framework designed to synthesize diverse experiences with scalability in mind to enable effective online RL training for autonomous agents. Rather than relying on expensive real-environment rollouts, DreamGym distills environment dynamics into a reasoning-based experience model that derives consistent state transitions and feedback signals through step-by-step reasoning, enabling scalable agent rollout collection for RL. To improve the stability and quality of transitions, DreamGym leverages an experience replay buffer initialized with offline real-world data and continuously enriched with fresh interactions to actively support agent training. To improve knowledge acquisition, DreamGym adaptively generates new tasks that challenge the current agent policy, enabling more effective online curriculum learning. Experiments across diverse environments and agent backbones demonstrate that DreamGym substantially improves RL training, both in fully synthetic settings and in sim-to-real transfer scenarios. On non-RL-ready tasks like WebArena, DreamGym outperforms all baselines by over 30%. And in RL-ready but costly settings, it matches GRPO and PPO performance using only synthetic interactions. When transferring a policy trained purely on synthetic experiences to real-environment RL, DreamGym yields significant additional performance gains while requiring far fewer real-world interactions, providing a scalable warm-start strategy for general-purpose RL. ---