# Research Papers — 2025-11 · bundle page 99 of 197 > Full-text markdown of papers #4901–4950 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/99 [← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/98.md) [Next bundle →](https://aipod.dev/papers/archive/2025-11/page/100.md) ### DIGing--SGLD: Decentralized and Scalable Langevin Sampling over Time--Varying Networks Authors: Waheed U. Bajwa, Mert Gurbuzbalaban, Mustafa Ali Kutbay, Lingjiong Zhu, Muhammad Zulqarnain Published: 2025-11-16 Categories: math.OC, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2511.12836v2 Sampling from a target distribution induced by training data is central to Bayesian learning, with Stochastic Gradient Langevin Dynamics (SGLD) serving as a key tool for scalable posterior sampling and decentralized variants enabling learning when data are distributed across a network of agents. This paper introduces DIGing-SGLD, a decentralized SGLD algorithm designed for scalable Bayesian learning in multi-agent systems operating over time-varying networks. Existing decentralized SGLD methods are restricted to static network topologies, and many exhibit steady-state sampling bias caused by network effects, even when full batches are used. DIGing-SGLD overcomes these limitations by integrating Langevin-based sampling with the gradient-tracking mechanism of the DIGing algorithm, originally developed for decentralized optimization over time-varying networks, thereby enabling efficient and bias-free sampling without a central coordinator. To our knowledge, we provide the first finite-time non-asymptotic Wasserstein convergence guarantees for decentralized SGLD-based sampling over time-varying networks, with explicit constants. Under standard strong convexity and smoothness assumptions, DIGing-SGLD achieves geometric convergence to an $O(\sqrtη)$ neighborhood of the target distribution, where $η$ is the stepsize, with dependence on the target accuracy matching the best-known rates for centralized and static-network SGLD algorithms using constant stepsize. Numerical experiments on Bayesian linear and logistic regression validate the theoretical results and demonstrate the strong empirical performance of DIGing-SGLD under dynamically evolving network conditions. --- ### SAGA: Source Attribution of Generative AI Videos Authors: Rohit Kundu, Vishal Mohanty, Hao Xiong, Shan Jia, Athula Balachandran, Amit K. Roy-Chowdhury Published: 2025-11-16 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.12834v2 The proliferation of generative AI has led to hyper-realistic synthetic videos, escalating misuse risks and outstripping binary real/fake detectors. We introduce SAGA (Source Attribution of Generative AI videos), the first comprehensive framework to address the urgent need for AI-generated video source attribution at a large scale. Unlike traditional detection, SAGA identifies the specific generative model used. It uniquely provides multi-granular attribution across five levels: authenticity, generation task (e.g., T2V/I2V), model version, development team, and the precise generator, offering far richer forensic insights. Our novel video transformer architecture, leveraging features from a robust vision foundation model, effectively captures spatio-temporal artifacts. Critically, we introduce a data-efficient pretrain-and-attribute strategy, enabling SAGA to achieve state-of-the-art attribution using only 0.5\% of source-labeled data per class, matching fully supervised performance. Furthermore, we propose Temporal Attention Signatures (T-Sigs), a novel interpretability method that visualizes learned temporal differences, offering the first explanation for why different video generators are distinguishable. Extensive experiments on public datasets, including cross-domain scenarios, demonstrate that SAGA sets a new benchmark for synthetic video provenance, providing crucial, interpretable insights for forensic and regulatory applications. --- ### From Passive to Persuasive: Localized Activation Injection for Empathy and Negotiation Authors: Niranjan Chebrolu, Kokil Jaidka, Gerard Christopher Yeo Published: 2025-11-16 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2511.12832v3 Complex social behaviors, such as empathy and strategic politeness, are widely assumed to resist the directional decomposition that makes activation steering effective for coarse attributes like sentiment or toxicity. We present STAR: Steering via Attribution and Representation, which tests this assumption by using attribution patching to identify the layer--token positions where each behavioral trait causally originates, then injecting contrastive activation vectors at precisely those locations. Evaluated on emotional dialogue and negotiation in both single- and multi-turn settings, localized injection consistently outperforms global steering and instruction priming; human evaluation confirms that gains reflect genuine improvements in perceived quality rather than lexical surface change. Our results suggest that complex interpersonal behaviors are encoded as localized, approximately linear directions in LLM activation space, and that behavioral alignment is fundamentally a localization problem. --- ### An Evaluation of Representation Learning Methods in Particle Physics Foundation Models Authors: Michael Chen, Raghav Kansal, Abhijith Gandrakota, Zichun Hao, Jennifer Ngadiuba, Maria Spiropulu Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12829v1 We present a systematic evaluation of representation learning objectives for particle physics within a unified framework. Our study employs a shared transformer-based particle-cloud encoder with standardized preprocessing, matched sampling, and a consistent evaluation protocol on a jet classification dataset. We compare contrastive (supervised and self-supervised), masked particle modeling, and generative reconstruction objectives under a common training regimen. In addition, we introduce targeted supervised architectural modifications that achieve state-of-the-art performance on benchmark evaluations. This controlled comparison isolates the contributions of the learning objective, highlights their respective strengths and limitations, and provides reproducible baselines. We position this work as a reference point for the future development of foundation models in particle physics, enabling more transparent and robust progress across the community. --- ### Catastrophic Forgetting in Kolmogorov-Arnold Networks Authors: Mohammad Marufur Rahman, Guanchu Wang, Kaixiong Zhou, Minghan Chen, Fan Yang Published: 2025-11-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.12828v1 Catastrophic forgetting is a longstanding challenge in continual learning, where models lose knowledge from earlier tasks when learning new ones. While various mitigation strategies have been proposed for Multi-Layer Perceptrons (MLPs), recent architectural advances like Kolmogorov-Arnold Networks (KANs) have been suggested to offer intrinsic resistance to forgetting by leveraging localized spline-based activations. However, the practical behavior of KANs under continual learning remains unclear, and their limitations are not well understood. To address this, we present a comprehensive study of catastrophic forgetting in KANs and develop a theoretical framework that links forgetting to activation support overlap and intrinsic data dimension. We validate these analyses through systematic experiments on synthetic and vision tasks, measuring forgetting dynamics under varying model configurations and data complexity. Further, we introduce KAN-LoRA, a novel adapter design for parameter-efficient continual fine-tuning of language models, and evaluate its effectiveness in knowledge editing tasks. Our findings reveal that while KANs exhibit promising retention in low-dimensional algorithmic settings, they remain vulnerable to forgetting in high-dimensional domains such as image classification and language modeling. These results advance the understanding of KANs' strengths and limitations, offering practical insights for continual learning system design. --- ### Efficient Adversarial Malware Defense via Trust-Based Raw Override and Confidence-Adaptive Bit-Depth Reduction Authors: Ayush Chaudhary, Sisir Doppalpudi Published: 2025-11-16 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2511.12827v1 The deployment of robust malware detection systems in big data environments requires careful consideration of both security effectiveness and computational efficiency. While recent advances in adversarial defenses have demonstrated strong robustness improvements, they often introduce computational overhead ranging from 4x to 22x, which presents significant challenges for production systems processing millions of samples daily. In this work, we propose a novel framework that combines Trust-Raw Override (TRO) with Confidence-Adaptive Bit-Depth Reduction (CABDR) to explicitly optimize the trade-off between adversarial robustness and computational efficiency. Our approach leverages adaptive confidence-based mechanisms to selectively apply defensive measures, achieving 1.76x computational overhead - a 2.3x improvement over state-of-the-art smoothing defenses. Through comprehensive evaluation on the EMBER v2 dataset comprising 800K samples, we demonstrate that our framework maintains 91 percent clean accuracy while reducing attack success rates to 31-37 percent across multiple attack types, with particularly strong performance against optimization-based attacks such as C and W (48.8 percent reduction). The framework achieves throughput of up to 1.26 million samples per second (measured on pre-extracted EMBER features with no runtime feature extraction), validated across 72 production configurations with statistical significance (5 independent runs, 95 percent confidence intervals, p less than 0.01). Our results suggest that practical adversarial robustness in production environments requires explicit optimization of the efficiency-robustness trade-off, providing a viable path for organizations to deploy robust defenses without prohibitive infrastructure costs. --- ### Enhancing LLM Code Generation Capabilities through Test-Driven Development and Code Interpreter Authors: Sajed Jalil, Shuvo Saha, Hossain Mohammad Seym Published: 2025-11-16 Categories: cs.SE, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2511.12823v1 Over the past few years, improving LLM code generation capabilities has been a key focus in NLP research. Despite Bengali having 242 million native speakers worldwide, it receives little attention when it comes to training LLMs. More recently, various fine-tuning and augmented generation techniques have been employed to significantly enhance code generation performance. However, they require considerable expertise and resources to utilize effectively as an end user. The goal of our work is to democratize access to powerful code generation tools in resource-constrained emerging markets, enabling users to leverage them in their native language. We introduce a novel approach that combines Test-Driven Development (TDD) and Code Interpreter (CI), utilizing open-weight models, which improves the baseline accuracy for code generation with Bengali prompts and achieves an overall accuracy of 85%. Our approach requires no finetuning and proves that even the smallest models in the same family can attain up to 98% accuracy compared to the largest models. All of our results are publicly shared in GitHub for validation and reproducibility. --- ### BioMedJImpact: A Comprehensive Dataset and LLM Pipeline for AI Engagement and Scientific Impact Analysis of Biomedical Journals Authors: Ruiyu Wang, Yuzhang Xie, Xiao Hu, Carl Yang, Jiaying Lu Published: 2025-11-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.12821v1 Assessing journal impact is central to scholarly communication, yet existing open resources rarely capture how collaboration structures and artificial intelligence (AI) research jointly shape venue prestige in biomedicine. We present BioMedJImpact, a large-scale, biomedical-oriented dataset designed to advance journal-level analysis of scientific impact and AI engagement. Built from 1.74 million PubMed Central articles across 2,744 journals, BioMedJImpact integrates bibliometric indicators, collaboration features, and LLM-derived semantic indicators for AI engagement. Specifically, the AI engagement feature is extracted through a reproducible three-stage LLM pipeline that we propose. Using this dataset, we analyze how collaboration intensity and AI engagement jointly influence scientific impact across pre- and post-pandemic periods (2016-2019, 2020-2023). Two consistent trends emerge: journals with higher collaboration intensity, particularly those with larger and more diverse author teams, tend to achieve greater citation impact, and AI engagement has become an increasingly strong correlate of journal prestige, especially in quartile rankings. To further validate the three-stage LLM pipeline we proposed for deriving the AI engagement feature, we conduct human evaluation, confirming substantial agreement in AI relevance detection and consistent subfield classification. Together, these contributions demonstrate that BioMedJImpact serves as both a comprehensive dataset capturing the intersection of biomedicine and AI, and a validated methodological framework enabling scalable, content-aware scientometric analysis of scientific impact and innovation dynamics. Code is available at https://github.com/JonathanWry/BioMedJImpact. --- ### Assessing Automated Fact-Checking for Medical LLM Responses with Knowledge Graphs Authors: Shasha Zhou, Mingyu Huang, Jack Cole, Charles Britton, Ming Yin, Jan Wolber, Ke Li Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12817v2 The recent proliferation of large language models (LLMs) holds the potential to revolutionize healthcare, with strong capabilities in diverse medical tasks. Yet, deploying LLMs in high-stakes healthcare settings requires rigorous verification and validation to understand any potential harm. This paper investigates the reliability and viability of using medical knowledge graphs (KGs) for the automated factuality evaluation of LLM-generated responses. To ground this investigation, we introduce FAITH, a framework designed to systematically probe the strengths and limitations of this KG-based approach. FAITH operates without reference answers by decomposing responses into atomic claims, linking them to a medical KG, and scoring them based on evidence paths. Experiments on diverse medical tasks with human subjective evaluations demonstrate that KG-grounded evaluation achieves considerably higher correlations with clinician judgments and can effectively distinguish LLMs with varying capabilities. It is also robust to textual variances. The inherent explainability of its scoring can further help users understand and mitigate the limitations of current LLMs. We conclude that while limitations exist, leveraging KGs is a prominent direction for automated factuality assessment in healthcare. --- ### MSRNet: A Multi-Scale Recursive Network for Camouflaged Object Detection Authors: Leena Alghamdi, Muhammad Usman, Hafeez Anwar, Abdul Bais, Saeed Anwar Published: 2025-11-16 Categories: cs.CV, cs.AI, eess.IV arXiv: https://arxiv.org/abs/2511.12810v2 Camouflaged object detection is an emerging and challenging computer vision task that requires identifying and segmenting objects that blend seamlessly into their environments due to high similarity in color, texture, and size. This task is further complicated by low-light conditions, partial occlusion, small object size, intricate background patterns, and multiple objects. While many sophisticated methods have been proposed for this task, current methods still struggle to precisely detect camouflaged objects in complex scenarios, especially with small and multiple objects, indicating room for improvement. We propose a Multi-Scale Recursive Network that extracts multi-scale features using a Pyramid Vision Transformer backbone and combines them with specialized Attention-Based Scale Integration Units, thereby enabling selective feature merging. For more precise object detection, our decoder recursively refines features by incorporating Multi-Granularity Fusion Units. A novel recursive-feedback decoding strategy is developed to enhance the model's understanding of global context, thereby helping it overcome the challenges of this task. By jointly leveraging multi-scale learning and recursive feature optimization, our proposed method achieves performance gains, successfully detecting small and multiple camouflaged objects. Our model achieves state-of-the-art results on two benchmark datasets for camouflaged object detection and ranks second on the remaining two. Our code, model weights, and results are available at https://github.com/linaagh98/MSRNet. --- ### Expressive Temporal Specifications for Reward Monitoring Authors: Omar Adalat, Francesco Belardinelli Published: 2025-11-16 Categories: cs.LG, cs.AI, cs.LO arXiv: https://arxiv.org/abs/2511.12808v4 Specifying informative and dense reward functions remains a pivotal challenge in Reinforcement Learning, as it directly affects the efficiency of agent training. In this work, we harness the expressive power of quantitative Linear Temporal Logic on finite traces (($\text{LTL}_f(\mathcal{F})$)) to synthesize reward monitors that generate a dense stream of rewards for runtime-observable state trajectories. By providing nuanced feedback during training, these monitors guide agents toward optimal behaviour and help mitigate the well-known issue of sparse rewards under long-horizon decision making, which arises under the Boolean semantics dominating the current literature. Our framework is algorithm-agnostic and only relies on a state labelling function, and naturally accommodates specifying non-Markovian properties. Empirical results show that our quantitative monitors consistently subsume and, depending on the environment, outperform Boolean monitors in maximizing a quantitative measure of task completion and in reducing convergence time. --- ### Practical Causal Evaluation Metrics for Biological Networks Authors: Noriaki Sato, Marco Scutari, Shuichi Kawano, Rui Yamaguchi, Seiya Imoto Published: 2025-11-16 Categories: q-bio.MN, cs.LG arXiv: https://arxiv.org/abs/2511.12805v1 Estimating causal networks from biological data is a critical step in systems biology. When evaluating the inferred network, assessing the networks based on their intervention effects is particularly important for downstream probabilistic reasoning and the identification of potential drug targets. In the context of gene regulatory network inference, biological databases are often used as reference sources. These databases typically describe relationships in a qualitative rather than quantitative manner. However, few evaluation metrics have been developed that take this qualitative nature into account. To address this, we developed a metric, the sign-augmented Structural Intervention Distance (sSID), and a weighted sSID that incorporates the net effects of the intervention. Through simulations and analyses of real transcriptomic datasets, we found that our proposed metrics could identify a different algorithm as optimal compared to conventional metrics, and the network selected by sSID had a superior performance in the classification task of clinical covariates using transcriptomic data. This suggests that sSID can distinguish networks that are structurally correct but functionally incorrect, highlighting its potential as a more biologically meaningful and practical evaluation metric. --- ### The Alignment Game: A Theory of Long-Horizon Alignment Through Recursive Curation Authors: Ali Falahati, Mohammad Mohammadi Amiri, Kate Larson, Lukasz Golab Published: 2025-11-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.12804v1 In self-consuming generative models that train on their own outputs, alignment with user preferences becomes a recursive rather than one-time process. We provide the first formal foundation for analyzing the long-term effects of such recursive retraining on alignment. Under a two-stage curation mechanism based on the Bradley-Terry (BT) model, we model alignment as an interaction between two factions: the Model Owner, who filters which outputs should be learned by the model, and the Public User, who determines which outputs are ultimately shared and retained through interactions with the model. Our analysis reveals three structural convergence regimes depending on the degree of preference alignment: consensus collapse, compromise on shared optima, and asymmetric refinement. We prove a fundamental impossibility theorem: no recursive BT-based curation mechanism can simultaneously preserve diversity, ensure symmetric influence, and eliminate dependence on initialization. Framing the process as dynamic social choice, we show that alignment is not a static goal but an evolving equilibrium, shaped both by power asymmetries and path dependence. --- ### Finite-Horizon Quickest Change Detection Balancing Latency with False Alarm Probability Authors: Yu-Han Huang, Venugopal V. Veeravalli Published: 2025-11-16 Categories: cs.IT, stat.ML arXiv: https://arxiv.org/abs/2511.12803v1 A finite-horizon variant of the quickest change detection (QCD) problem that is of relevance to learning in non-stationary environments is studied. The metric characterizing false alarms is the probability of a false alarm occurring before the horizon ends. The metric that characterizes the delay is \emph{latency}, which is the smallest value such that the probability that detection delay exceeds this value is upper bounded to a predetermined latency level. The objective is to minimize the latency (at a given latency level), while maintaining a low false alarm probability. Under the pre-specified latency and false alarm levels, a universal lower bound on the latency, which any change detection procedure needs to satisfy, is derived. Change detectors are then developed, which are order-optimal in terms of the horizon. The case where the pre- and post-change distributions are known is considered first, and then the results are generalized to the non-parametric case when they are unknown except that they are sub-Gaussian with different means. Simulations are provided to validate the theoretical results. --- ### Enhancing Neuro-Oncology Through Self-Assessing Deep Learning Models for Brain Tumor Unified Model for MRI Segmentation Authors: Andrew Zhou Published: 2025-11-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.12801v1 Accurate segmentation of brain tumors is vital for diagnosis, surgical planning, and treatment monitoring. Deep learning has advanced on benchmarks, but two issues limit clinical use: no uncertainty estimates for errors and no segmentation of healthy brain structures around tumors for surgery. Current methods fail to unify tumor localization with anatomical context and lack confidence scores. This study presents an uncertainty-aware framework augmenting nnUNet with a channel for voxel-wise uncertainty. Trained on BraTS2023, it yields a correlation of 0.750 and RMSD of 0.047 for uncertainty without hurting tumor accuracy. It predicts uncertainty in one pass, with no extra networks or inferences, aiding clinical decisions. For whole-brain context, a unified model combines normal and cancer datasets, achieving a DSC of 0.81 for brain structures and 0.86 for tumor, with robust key-region performance. Combining both innovations gives the first model outputting tumor in natural surroundings plus an overlaid uncertainty map. Visual checks of outputs show uncertainty offers key insights to evaluate predictions and fix errors, helping informed surgical decisions from AI. --- ### Genomic Next-Token Predictors are In-Context Learners Authors: Nathan Breslow, Aayush Mishra, Mahler Revsine, Michael C. Schatz, Anqi Liu, Daniel Khashabi Published: 2025-11-16 Categories: cs.LG, cs.AI, q-bio.GN arXiv: https://arxiv.org/abs/2511.12797v3 In-context learning (ICL) -- the capacity of a model to infer and apply abstract patterns from examples provided within its input -- has been extensively studied in large language models trained for next-token prediction on human text. In fact, prior work often attributes this emergent behavior to distinctive statistical properties in human language. This raises a fundamental question: can ICL arise organically in other sequence domains purely through large-scale predictive training? To explore this, we turn to genomic sequences, an alternative symbolic domain rich in statistical structure. Specifically, we study the Evo2 genomic model, trained predominantly on next-nucleotide (A/T/C/G) prediction, at a scale comparable to mid-sized LLMs. We develop a controlled experimental framework comprising symbolic reasoning tasks instantiated in both linguistic and genomic forms, enabling direct comparison of ICL across genomic and linguistic models. Our results show that genomic models, like their linguistic counterparts, exhibit log-linear gains in pattern induction as the number of in-context demonstrations increases. To the best of our knowledge, this is the first evidence of organically emergent ICL in genomic sequences, supporting the hypothesis that ICL arises as a consequence of large-scale predictive modeling over rich data. These findings extend emergent meta-learning beyond language, pointing toward a unified, modality-agnostic view of in-context learning. --- ### Maximizing the efficiency of human feedback in AI alignment: a comparative analysis Authors: Andreas Chouliaras, Dimitris Chatzopoulos Published: 2025-11-16 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2511.12796v2 Reinforcement Learning from Human Feedback (RLHF) relies on preference modeling to align machine learning systems with human values, yet the popular approach of random pair sampling with Bradley-Terry modeling is statistically limited and inefficient under constrained annotation budgets. In this work, we explore alternative sampling and evaluation strategies for preference inference in RLHF, drawing inspiration from areas such as game theory, statistics, and social choice theory. Our best-performing method, Swiss InfoGain, employs a Swiss tournament system with a proxy mutual-information-gain pairing rule, which significantly outperforms all other methods in constrained annotation budgets while also being more sample-efficient. Even in high-resource settings, we can identify superior alternatives to the Bradley-Terry baseline. Our experiments demonstrate that adaptive, resource-aware strategies reduce redundancy, enhance robustness, and yield statistically significant improvements in preference learning, highlighting the importance of balancing alignment quality with human workload in RLHF pipelines. --- ### Neuro-Logic Lifelong Learning Authors: Bowen He, Xiaoan Xu, Alper Kamil Bozkurt, Vahid Tarokh, Juncheng Dong Published: 2025-11-16 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.12793v1 Solving Inductive Logic Programming (ILP) problems with neural networks is a key challenge in Neural-Symbolic Ar- tificial Intelligence (AI). While most research has focused on designing novel network architectures for individual prob- lems, less effort has been devoted to exploring new learning paradigms involving a sequence of problems. In this work, we investigate lifelong learning ILP, which leverages the com- positional and transferable nature of logic rules for efficient learning of new problems. We introduce a compositional framework, demonstrating how logic rules acquired from ear- lier tasks can be efficiently reused in subsequent ones, leading to improved scalability and performance. We formalize our approach and empirically evaluate it on sequences of tasks. Experimental results validate the feasibility and advantages of this paradigm, opening new directions for continual learn- ing in Neural-Symbolic AI. --- ### Multi-Agent Reinforcement Learning for Heterogeneous Satellite Cluster Resources Optimization Authors: Mohamad A. Hady, Siyi Hu, Mahardhika Pratama, Zehong Cao, Ryszard Kowalczyk Published: 2025-11-16 Categories: cs.AI arXiv: https://arxiv.org/abs/2511.12792v1 This work investigates resource optimization in heterogeneous satellite clusters performing autonomous Earth Observation (EO) missions using Reinforcement Learning (RL). In the proposed setting, two optical satellites and one Synthetic Aperture Radar (SAR) satellite operate cooperatively in low Earth orbit to capture ground targets and manage their limited onboard resources efficiently. Traditional optimization methods struggle to handle the real-time, uncertain, and decentralized nature of EO operations, motivating the use of RL and Multi-Agent Reinforcement Learning (MARL) for adaptive decision-making. This study systematically formulates the optimization problem from single-satellite to multi-satellite scenarios, addressing key challenges including energy and memory constraints, partial observability, and agent heterogeneity arising from diverse payload capabilities. Using a near-realistic simulation environment built on the Basilisk and BSK-RL frameworks, we evaluate the performance and stability of state-of-the-art MARL algorithms such as MAPPO, HAPPO, and HATRPO. Results show that MARL enables effective coordination across heterogeneous satellites, balancing imaging performance and resource utilization while mitigating non-stationarity and inter-agent reward coupling. The findings provide practical insights into scalable, autonomous satellite operations and contribute a foundation for future research on intelligent EO mission planning under heterogeneous and dynamic conditions. --- ### Optimal Look-back Horizon for Time Series Forecasting in Federated Learning Authors: Dahao Tang, Nan Yang, Yanli Li, Zhiyu Zhu, Zhibo Jin, Dong Yuan Published: 2025-11-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.12791v3 Selecting an appropriate look-back horizon remains a fundamental challenge in time series forecasting (TSF), particularly in the federated learning scenarios where data is decentralized, heterogeneous, and often non-independent. While recent work has explored horizon selection by preserving forecasting-relevant information in an intrinsic space, these approaches are primarily restricted to centralized and independently distributed settings. This paper presents a principled framework for adaptive horizon selection in federated time series forecasting through an intrinsic space formulation. We introduce a synthetic data generator (SDG) that captures essential temporal structures in client data, including autoregressive dependencies, seasonality, and trend, while incorporating client-specific heterogeneity. Building on this model, we define a transformation that maps time series windows into an intrinsic representation space with well-defined geometric and statistical properties. We then derive a decomposition of the forecasting loss into a Bayesian term, which reflects irreducible uncertainty, and an approximation term, which accounts for finite-sample effects and limited model capacity. Our analysis shows that while increasing the look-back horizon improves the identifiability of deterministic patterns, it also increases approximation error due to higher model complexity and reduced sample efficiency. We prove that the total forecasting loss is minimized at the smallest horizon where the irreducible loss starts to saturate, while the approximation loss continues to rise. This work provides a rigorous theoretical foundation for adaptive horizon selection for time series forecasting in federated learning. --- ### Physics-Constrained Adaptive Neural Networks Enable Real-Time Semiconductor Manufacturing Optimization with Minimal Training Data Authors: Rubén Darío Guerrero Published: 2025-11-16 Categories: cs.LG, cs.AR, math.OC arXiv: https://arxiv.org/abs/2511.12788v1 The semiconductor industry faces a computational crisis in extreme ultraviolet (EUV) lithography optimization, where traditional methods consume billions of CPU hours while failing to achieve sub-nanometer precision. We present a physics-constrained adaptive learning framework that automatically calibrates electromagnetic approximations through learnable parameters $\boldsymbolθ = \{θ_d, θ_a, θ_b, θ_p, θ_c\}$ while simultaneously minimizing Edge Placement Error (EPE) between simulated aerial images and target photomasks. The framework integrates differentiable modules for Fresnel diffraction, material absorption, optical point spread function blur, phase-shift effects, and contrast modulation with direct geometric pattern matching objectives, enabling cross-geometry generalization with minimal training data. Through physics-constrained learning on 15 representative patterns spanning current production to future research nodes, we demonstrate consistent sub-nanometer EPE performance (0.664-2.536 nm range) using only 50 training samples per pattern. Adaptive physics learning achieves an average improvement of 69.9\% over CNN baselines without physics constraints, with a significant inference speedup over rigorous electromagnetic solvers after training completion. This approach requires 90\% fewer training samples through cross-geometry generalization compared to pattern-specific CNN training approaches. This work establishes physics-constrained adaptive learning as a foundational methodology for real-time semiconductor manufacturing optimization, addressing the critical gap between academic physics-informed neural networks and industrial deployment requirements through joint physics calibration and manufacturing precision objectives. --- ### Lightweight Optimal-Transport Harmonization on Edge Devices Authors: Maria Larchenko, Dmitry Guskov, Alexander Lobashev, Georgy Derevyanko Published: 2025-11-16 Categories: cs.CV, cs.AI, cs.GR arXiv: https://arxiv.org/abs/2511.12785v1 Color harmonization adjusts the colors of an inserted object so that it perceptually matches the surrounding image, resulting in a seamless composite. The harmonization problem naturally arises in augmented reality (AR), yet harmonization algorithms are not currently integrated into AR pipelines because real-time solutions are scarce. In this work, we address color harmonization for AR by proposing a lightweight approach that supports on-device inference. For this, we leverage classical optimal transport theory by training a compact encoder to predict the Monge-Kantorovich transport map. We benchmark our MKL-Harmonizer algorithm against state-of-the-art methods and demonstrate that for real composite AR images our method achieves the best aggregated score. We release our dedicated AR dataset of composite images with pixel-accurate masks and data-gathering toolkit to support further data acquisition by researchers. --- ### Evaluating Autoformalization Robustness via Semantically Similar Paraphrasing Authors: Hayden Moore, Asfahan Shah Published: 2025-11-16 Categories: cs.CL, cs.LO arXiv: https://arxiv.org/abs/2511.12784v3 Large Language Models (LLMs) have recently emerged as powerful tools for autoformalization. Despite their impressive performance, these models can still struggle to produce grounded and verifiable formalizations. Recent work in text-to-SQL, has revealed that LLMs can be sensitive to paraphrased natural language (NL) inputs, even when high degrees of semantic fidelity are preserved. In this paper, we investigate this claim in the autoformalization domain. Specifically, we evaluate the robustness of LLMs generating formal proofs with semantically similar paraphrased NL statements by measuring semantic and compilation validity. Using the formal benchmarks MiniF2F and Lean 4 version of ProofNet, and two modern LLMs, we generate paraphrased natural language statements and cross-evaluate these statements across both models. The results of this paper reveal performance variability across paraphrased inputs, demonstrating that minor shifts in NL statements can significantly impact model outputs. --- ### Function-on-Function Bayesian Optimization Authors: Jingru Huang, Haijie Xu, Manrui Jiang, Chen Zhang Published: 2025-11-16 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2511.12783v2 Bayesian optimization (BO) has been widely used to optimize expensive and gradient-free objective functions across various domains. However, existing BO methods have not addressed the objective where both inputs and outputs are functions, which increasingly arise in complex systems as advanced sensing technologies. To fill this gap, we propose a novel function-on-function Bayesian optimization (FFBO) framework. Specifically, we first introduce a function-on-function Gaussian process (FFGP) model with a separable operator-valued kernel to capture the correlations between function-valued inputs and outputs. Compared to existing Gaussian process models, FFGP is modeled directly in the function space. Based on FFGP, we define a scalar upper confidence bound (UCB) acquisition function using a weighted operator-based scalarization strategy. Then, a scalable functional gradient ascent algorithm (FGA) is developed to efficiently identify the optimal function-valued input. We further analyze the theoretical properties of the proposed method. Extensive experiments on synthetic and real-world data demonstrate the superior performance of FFBO over existing approaches. --- ### LLM Reinforcement in Context Authors: Thomas Rivasseau Published: 2025-11-16 Categories: cs.CL, cs.CR arXiv: https://arxiv.org/abs/2511.12782v1 Current Large Language Model alignment research mostly focuses on improving model robustness against adversarial attacks and misbehavior by training on examples and prompting. Research has shown that LLM jailbreak probability increases with the size of the user input or conversation length. There is a lack of appropriate research into means of strengthening alignment which also scale with user input length. We propose interruptions as a possible solution to this problem. Interruptions are control sentences added to the user input approximately every x tokens for some arbitrary x. We suggest that this can be generalized to the Chain-of-Thought process to prevent scheming. --- ### Scalable Multi-Objective and Meta Reinforcement Learning via Gradient Estimation Authors: Zhenshuo Zhang, Minxuan Duan, Youran Ye, Hongyang R. Zhang Published: 2025-11-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.12779v3 We study the problem of efficiently estimating policies that simultaneously optimize multiple objectives in reinforcement learning (RL). Given $n$ objectives (or tasks), we seek the optimal partition of these objectives into $k \ll n$ groups, where each group comprises related objectives that can be trained together. This problem arises in applications such as robotics, control, and preference optimization in language models, where learning a single policy for all $n$ objectives is suboptimal as $n$ grows. We introduce a two-stage procedure -- meta-training followed by fine-tuning -- to address this problem. We first learn a meta-policy for all objectives using multitask learning. Then, we adapt the meta-policy to multiple randomly sampled subsets of objectives. The adaptation step leverages a first-order approximation property of well-trained policy networks, which is empirically verified to be accurate within a 2% error margin across various RL environments. The resulting algorithm, PolicyGradEx, efficiently estimates an aggregate task-affinity score matrix given a policy evaluation algorithm. Based on the estimated affinity score matrix, we cluster the $n$ objectives into $k$ groups by maximizing the intra-cluster affinity scores. Experiments on three robotic control and the Meta-World benchmarks demonstrate that our approach outperforms state-of-the-art baselines by 16% on average, while delivering up to $26\times$ faster speedup relative to performing full training to obtain the clusters. Ablation studies validate each component of our approach. For instance, compared with random grouping and gradient-similarity-based grouping, our loss-based clustering yields an improvement of 19%. Finally, we analyze the generalization error of policy networks by measuring the Hessian trace of the loss surface, which gives non-vacuous measures relative to the observed generalization errors. --- ### MolEdit: Knowledge Editing for Multimodal Molecule Language Models Authors: Zhenyu Lei, Patrick Soga, Yaochen Zhu, Yinhan He, Yushun Dong, Jundong Li Published: 2025-11-16 Categories: cs.LG, cs.CE arXiv: https://arxiv.org/abs/2511.12770v2 Understanding and continuously refining multimodal molecular knowledge is crucial for advancing biomedicine, chemistry, and materials science. Molecule language models (MoLMs) have become powerful tools in these domains, integrating structural representations (e.g., SMILES strings, molecular graphs) with rich contextual descriptions (e.g., physicochemical properties). However, MoLMs can encode and propagate inaccuracies due to outdated web-mined training corpora or malicious manipulation, jeopardizing downstream discovery pipelines. While knowledge editing has been explored for general-domain AI, its application to MoLMs remains uncharted, presenting unique challenges due to the multifaceted and interdependent nature of molecular knowledge. In this paper, we take the first step toward MoLM editing for two critical tasks: molecule-to-caption generation and caption-to-molecule generation. To address molecule-specific challenges, we propose MolEdit, a powerful framework that enables targeted modifications while preserving unrelated molecular knowledge. MolEdit combines a Multi-Expert Knowledge Adapter that routes edits to specialized experts for different molecular facets with an Expertise-Aware Editing Switcher that activates the adapters only when input closely matches the stored edits across all expertise, minimizing interference with unrelated knowledge. To systematically evaluate editing performance, we introduce MEBench, a comprehensive benchmark assessing multiple dimensions, including Reliability (accuracy of the editing), Locality (preservation of irrelevant knowledge), and Generality (robustness to reformed queries). Across extensive experiments on two popular MoLM backbones, MolEdit delivers up to 18.8% higher Reliability and 12.0% better Locality than baselines while maintaining efficiency. The code is available at: https://github.com/LzyFischer/MolEdit. --- ### Event-CausNet: Unlocking Causal Knowledge from Text with Large Language Models for Reliable Spatio-Temporal Forecasting Authors: Luyao Niu, Zepu Wang, Shuyi Guan, Yang Liu, Peng Sun Published: 2025-11-16 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.12769v1 While spatio-temporal Graph Neural Networks (GNNs) excel at modeling recurring traffic patterns, their reliability plummets during non-recurring events like accidents. This failure occurs because GNNs are fundamentally correlational models, learning historical patterns that are invalidated by the new causal factors introduced during disruptions. To address this, we propose Event-CausNet, a framework that uses a Large Language Model to quantify unstructured event reports, builds a causal knowledge base by estimating average treatment effects, and injects this knowledge into a dual-stream GNN-LSTM network using a novel causal attention mechanism to adjust and enhance the forecast. Experiments on a real-world dataset demonstrate that Event-CausNet achieves robust performance, reducing prediction error (MAE) by up to 35.87%, significantly outperforming state-of-the-art baselines. Our framework bridges the gap between correlational models and causal reasoning, providing a solution that is more accurate and transferable, while also offering crucial interpretability, providing a more reliable foundation for real-world traffic management during critical disruptions. --- ### Evidence of Phase Transitions in Small Transformer-Based Language Models Authors: Noah Hong, Tao Hong Published: 2025-11-16 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2511.12768v1 Phase transitions have been proposed as the origin of emergent abilities in large language models (LLMs), where new capabilities appear abruptly once models surpass critical thresholds of scale. Prior work, such as that of Wei et al., demonstrated these phenomena under model and data scaling, with transitions revealed after applying a log scale to training compute. In this work, we ask three complementary questions: (1) Are phase transitions unique to large models, or can they also be observed in small transformer-based language models? (2) Can such transitions be detected directly in linear training space, rather than only after log rescaling? and (3) Can these transitions emerge at early stages of training? To investigate, we train a small GPT-style transformer on a character-level corpus and analyze the evolution of vocabulary usage throughout training. We track the average word length, the number of correct versus incorrect words, and shifts in vocabulary diversity. Building on these measures, we apply Poisson and sub-Poisson statistics to quantify how words connect and reorganize. This combined analysis reveals a distinct transition point during training. Notably, these transitions are not apparent in standard loss or validation curves, but become visible through our vocabulary- and statistics-based probes. Our findings suggest that phase-transition reorganizations are a general feature of language model training, observable even in modest models, detectable directly in linear training space, and occurring surprisingly early as coherence emerges. This perspective provides new insight into the nonlinear dynamics of language model training and underscores the importance of tailored metrics for uncovering phase transition behaviors --- ### RoCoISLR: A Romanian Corpus for Isolated Sign Language Recognition Authors: Cătălin-Alexandru Rîpanu, Andrei-Theodor Hotnog, Giulia-Stefania Imbrea, Dumitru-Clementin Cercel Published: 2025-11-16 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.12767v1 Automatic sign language recognition plays a crucial role in bridging the communication gap between deaf communities and hearing individuals; however, most available datasets focus on American Sign Language. For Romanian Isolated Sign Language Recognition (RoISLR), no large-scale, standardized dataset exists, which limits research progress. In this work, we introduce a new corpus for RoISLR, named RoCoISLR, comprising over 9,000 video samples that span nearly 6,000 standardized glosses from multiple sources. We establish benchmark results by evaluating seven state-of-the-art video recognition models-I3D, SlowFast, Swin Transformer, TimeSformer, Uniformer, VideoMAE, and PoseConv3D-under consistent experimental setups, and compare their performance with that of the widely used WLASL2000 corpus. According to the results, transformer-based architectures outperform convolutional baselines; Swin Transformer achieved a Top-1 accuracy of 34.1%. Our benchmarks highlight the challenges associated with long-tail class distributions in low-resource sign languages, and RoCoISLR provides the initial foundation for systematic RoISLR research. --- ### INC: An Indirect Neural Corrector for Auto-Regressive Hybrid PDE Solvers Authors: Hao Wei, Aleksandra Franz, Bjoern List, Nils Thuerey Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12764v2 When simulating partial differential equations, hybrid solvers combine coarse numerical solvers with learned correctors. They promise accelerated simulations while adhering to physical constraints. However, as shown in our theoretical framework, directly applying learned corrections to solver outputs leads to significant autoregressive errors, which originate from amplified perturbations that accumulate during long-term rollouts, especially in chaotic regimes. To overcome this, we propose the Indirect Neural Corrector ($\mathrm{INC}$), which integrates learned corrections into the governing equations rather than applying direct state updates. Our key insight is that $\mathrm{INC}$ reduces the error amplification on the order of $Δt^{-1} + L$, where $Δt$ is the timestep and $L$ the Lipschitz constant. At the same time, our framework poses no architectural requirements and integrates seamlessly with arbitrary neural networks and solvers. We test $\mathrm{INC}$ in extensive benchmarks, covering numerous differentiable solvers, neural backbones, and test cases ranging from a 1D chaotic system to 3D turbulence. $\mathrm{INC}$ improves the long-term trajectory performance ($R^2$) by up to 158.7%, stabilizes blowups under aggressive coarsening, and for complex 3D turbulence cases yields speed-ups of several orders of magnitude. $\mathrm{INC}$ thus enables stable, efficient PDE emulation with formal error reduction, paving the way for faster scientific and engineering simulations with reliable physics guarantees. Our source code is available at https://github.com/tum-pbs/INC --- ### Conformal Online Learning of Deep Koopman Linear Embeddings Authors: Ben Gao, Jordan Patracone, Stéphane Chrétien, Olivier Alata Published: 2025-11-16 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2511.12760v2 We introduce Conformal Online Learning of Koopman embeddings (COLoKe), a novel framework for adaptively updating Koopman-invariant representations of nonlinear dynamical systems from streaming data. Our modeling approach combines deep feature learning with multistep prediction consistency in the lifted space, where the dynamics evolve linearly. To prevent overfitting, COLoKe employs a conformal-style mechanism that shifts the focus from evaluating the conformity of new states to assessing the consistency of the current Koopman model. Updates are triggered only when the current model's prediction error exceeds a dynamically calibrated threshold, allowing selective refinement of the Koopman operator and embedding. Empirical results on benchmark dynamical systems demonstrate the effectiveness of COLoKe in maintaining long-term predictive accuracy while significantly reducing unnecessary updates and avoiding overfitting. --- ### Optimal Foraging in Memory Retrieval: Evaluating Random Walks and Metropolis-Hastings Sampling in Modern Semantic Spaces Authors: James Moore Published: 2025-11-16 Categories: cs.AI arXiv: https://arxiv.org/abs/2511.12759v1 Human memory retrieval often resembles ecological foraging where animals search for food in a patchy environment. Optimal foraging means following the Marginal Value Theorem (MVT), in which individuals exploit a patch of semantically related concepts until it becomes less rewarding and then switch to a new cluster. While human behavioral data suggests foraging-like patterns in semantic fluency tasks, it remains unclear whether modern high-dimensional embedding spaces provide representations that allow algorithms to match observed human behavior. Using state-of-the-art embeddings and prior semantic fluency data, I find that random walks on these embedding spaces produce results consistent with optimal foraging and the MVT. Surprisingly, introducing Metropolis-Hastings sampling, an adaptive algorithm expected to model strategic acceptance and rejection of new clusters, does not produce results consistent with human behavior. These findings challenge the assumption that more complex sampling mechanisms inherently lead to better cognitive models of memory retrieval. Instead, they show that appropriately structured embeddings, even with simple sampling, can produce near-optimal foraging dynamics. This supports the perspective of Hills (2012) rather than Abbott (2015), demonstrating that modern embeddings can approximate human memory foraging without relying on complex acceptance criteria. --- ### Which Way from B to A: The role of embedding geometry in image interpolation for Stable Diffusion Authors: Nicholas Karris, Luke Durell, Javier Flores, Tegan Emerson Published: 2025-11-16 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.12757v1 It can be shown that Stable Diffusion has a permutation-invariance property with respect to the rows of Contrastive Language-Image Pretraining (CLIP) embedding matrices. This inspired the novel observation that these embeddings can naturally be interpreted as point clouds in a Wasserstein space rather than as matrices in a Euclidean space. This perspective opens up new possibilities for understanding the geometry of embedding space. For example, when interpolating between embeddings of two distinct prompts, we propose reframing the interpolation problem as an optimal transport problem. By solving this optimal transport problem, we compute a shortest path (or geodesic) between embeddings that captures a more natural and geometrically smooth transition through the embedding space. This results in smoother and more coherent intermediate (interpolated) images when rendered by the Stable Diffusion generative model. We conduct experiments to investigate this effect, comparing the quality of interpolated images produced using optimal transport to those generated by other standard interpolation methods. The novel optimal transport--based approach presented indeed gives smoother image interpolations, suggesting that viewing the embeddings as point clouds (rather than as matrices) better reflects and leverages the geometry of the embedding space. --- ### Prompt-Driven Domain Adaptation for End-to-End Autonomous Driving via In-Context RL Authors: Aleesha Khurram, Amir Moeini, Shangtong Zhang, Rohan Chandra Published: 2025-11-16 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2511.12755v1 Despite significant progress and advances in autonomous driving, many end-to-end systems still struggle with domain adaptation (DA), such as transferring a policy trained under clear weather to adverse weather conditions. Typical DA strategies in the literature include collecting additional data in the target domain or re-training the model, or both. Both these strategies quickly become impractical as we increase scale and complexity of driving. These limitations have encouraged investigation into few-shot and zero-shot prompt-driven DA at inference time involving LLMs and VLMs. These methods work by adding a few state-action trajectories during inference to the prompt (similar to in-context learning). However, there are two limitations of such an approach: $(i)$ prompt-driven DA methods are currently restricted to perception tasks such as detection and segmentation and $(ii)$ they require expert few-shot data. In this work, we present a new approach to inference-time few-shot prompt-driven DA for closed-loop autonomous driving in adverse weather condition using in-context reinforcement learning (ICRL). Similar to other prompt-driven DA methods, our approach does not require any updates to the model parameters nor does it require additional data collection in adversarial weather regime. Furthermore, our approach advances the state-of-the-art in prompt-driven DA by extending to closed driving using general trajectories observed during inference. Our experiments using the CARLA simulator show that ICRL results in safer, more efficient, and more comfortable driving policies in the target domain compared to state-of-the-art prompt-driven DA baselines. --- ### Adaptively Coordinating with Novel Partners via Learned Latent Strategies Authors: Benjamin Li, Shuyang Shi, Lucia Romero, Huao Li, Yaqi Xie, Woojun Kim, Stefanos Nikolaidis, Michael Lewis, Katia Sycara, Simon Stepputtis Published: 2025-11-16 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2511.12754v1 Adaptation is the cornerstone of effective collaboration among heterogeneous team members. In human-agent teams, artificial agents need to adapt to their human partners in real time, as individuals often have unique preferences and policies that may change dynamically throughout interactions. This becomes particularly challenging in tasks with time pressure and complex strategic spaces, where identifying partner behaviors and selecting suitable responses is difficult. In this work, we introduce a strategy-conditioned cooperator framework that learns to represent, categorize, and adapt to a broad range of potential partner strategies in real-time. Our approach encodes strategies with a variational autoencoder to learn a latent strategy space from agent trajectory data, identifies distinct strategy types through clustering, and trains a cooperator agent conditioned on these clusters by generating partners of each strategy type. For online adaptation to novel partners, we leverage a fixed-share regret minimization algorithm that dynamically infers and adjusts the partner's strategy estimation during interaction. We evaluate our method in a modified version of the Overcooked domain, a complex collaborative cooking environment that requires effective coordination among two players with a diverse potential strategy space. Through these experiments and an online user study, we demonstrate that our proposed agent achieves state of the art performance compared to existing baselines when paired with novel human, and agent teammates. --- ### Whose Narrative is it Anyway? A KV Cache Manipulation Attack Authors: Mukkesh Ganesh, Kaushik Iyer, Arun Baalaaji Sankar Ananthan Published: 2025-11-16 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2511.12752v1 The Key Value(KV) cache is an important component for efficient inference in autoregressive Large Language Models (LLMs), but its role as a representation of the model's internal state makes it a potential target for integrity attacks. This paper introduces "History Swapping," a novel block-level attack that manipulates the KV cache to steer model generation without altering the user-facing prompt. The attack involves overwriting a contiguous segment of the active generation's cache with a precomputed cache from a different topic. We empirically evaluate this method across 324 configurations on the Qwen 3 family of models, analyzing the impact of timing, magnitude, and layer depth of the cache overwrite. Our findings reveal that only full-layer overwrites can successfully hijack the conversation's topic, leading to three distinct behaviors: immediate and persistent topic shift, partial recovery, or a delayed hijack. Furthermore, we observe that high-level structural plans are encoded early in the generation process and local discourse structure is maintained by the final layers of the model. This work demonstrates that the KV cache is a significant vector for security analysis, as it encodes not just context but also topic trajectory and structural planning, making it a powerful interface for manipulating model behavior. --- ### Are LLMs The Way Forward? A Case Study on LLM-Guided Reinforcement Learning for Decentralized Autonomous Driving Authors: Timur Anvar, Jeffrey Chen, Yuyan Wang, Rohan Chandra Published: 2025-11-16 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2511.12751v1 Autonomous vehicle navigation in complex environments such as dense and fast-moving highways and merging scenarios remains an active area of research. A key limitation of RL is its reliance on well-specified reward functions, which often fail to capture the full semantic and social complexity of diverse, out-of-distribution situations. As a result, a rapidly growing line of research explores using Large Language Models (LLMs) to replace or supplement RL for direct planning and control, on account of their ability to reason about rich semantic context. However, LLMs present significant drawbacks: they can be unstable in zero-shot safety-critical settings, produce inconsistent outputs, and often depend on expensive API calls with network latency. This motivates our investigation into whether small, locally deployed LLMs (< 14B parameters) can meaningfully support autonomous highway driving through reward shaping rather than direct control. We present a case study comparing RL-only, LLM-only, and hybrid approaches, where LLMs augment RL rewards by scoring state-action transitions during training, while standard RL policies execute at test time. Our findings reveal that RL-only agents achieve moderate success rates (73-89%) with reasonable efficiency, LLM-only agents can reach higher success rates (up to 94%) but with severely degraded speed performance, and hybrid approaches consistently fall between these extremes. Critically, despite explicit efficiency instructions, LLM-influenced approaches exhibit systematic conservative bias with substantial model-dependent variability, highlighting important limitations of current small LLMs for safety-critical control tasks. --- ### When Does Pooling Pay? Credibility and Resolution under Forgetting in Intermittent-Demand Forecasting Authors: Zong-Han Bai, Po-Yen Chu Published: 2025-11-16 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2511.12749v3 Forecasting many sparse series requires two choices: how much of each series' own past to retain, and how much to borrow from other series. We show that when one exponential recency operator is applied to item- and group-level statistics alike in a hierarchical empirical-Bayes hurdle model, the two become one decision: forgetting preserves the leverage of a shared prior while raising the noise floor against which fine cross-series structure must be resolved. We formalize this two-sided effect as a fitting-window diagnostic, a credibility bound and a resolution condition for candidate pools, and derive from it an ex ante screen for regimes where refinement cannot pay. On five public intermittent-demand panels comprising over 19,000 series, where item-level evidence is sparsest, the diagnostic is right in both directions. Where truncation lowers credibility the shared prior pays: switching it off costs $5\%$ and $12\%$ at the shortest histories on the two panels it helps most, against about $1\%$ at full length, and the gain sits in the occurrence block. On four panels the diagnostic reports at most $0.6\%$ of room for refinement and realized gains are no larger; on the one panel with room the learned mixture does not collect it, which locates the open problem in the partition objective. The resulting forecaster, EBB, stays within about $3\%$ of a per-series Gaussian-process specialist at fixed origin and within $1\%$ under walk-forward evaluation where both run, at two to three orders of magnitude lower cost. --- ### DIVIDE: A Framework for Learning from Independent Multi-Mechanism Data Using Deep Encoders and Gaussian Processes Authors: Vivek Chawla, Boris Slautin, Utkarsh Pratiush, Dayakar Penumadu, Sergei Kalinin Published: 2025-11-16 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2511.12745v1 Scientific datasets often arise from multiple independent mechanisms such as spatial, categorical or structural effects, whose combined influence obscures their individual contributions. We introduce DIVIDE, a framework that disentangles these influences by integrating mechanism-specific deep encoders with a structured Gaussian Process in a joint latent space. Disentanglement here refers to separating independently acting generative factors. The encoders isolate distinct mechanisms while the Gaussian Process captures their combined effect with calibrated uncertainty. The architecture supports structured priors, enabling interpretable and mechanism-aware prediction as well as efficient active learning. DIVIDE is demonstrated on synthetic datasets combining categorical image patches with nonlinear spatial fields, on FerroSIM spin lattice simulations of ferroelectric patterns, and on experimental PFM hysteresis loops from PbTiO3 films. Across benchmarks, DIVIDE separates mechanisms, reproduces additive and scaled interactions, and remains robust under noise. The framework extends naturally to multifunctional datasets where mechanical, electromagnetic or optical responses coexist. --- ### SAGE: Saliency-Guided Contrastive Embeddings Authors: Colton R. Crum, Christopher Sweet, Adam Czajka Published: 2025-11-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.12744v2 Integrating human perceptual priors into the training of neural networks has been shown to raise model generalization, serve as an effective regularizer, and align models with human expertise for applications in high-risk domains. Existing approaches to integrate saliency into model training often rely on internal model mechanisms, which recent research suggests may be unreliable. Our insight is that many challenges associated with saliency-guided training stem from the placement of the guidance approaches solely within the image space. Instead, we move away from the image space, use the model's latent space embeddings to steer human guidance during training, and we propose SAGE (Saliency-Guided Contrastive Embeddings): a loss function that integrates human saliency into network training using contrastive embeddings. We apply salient-preserving and saliency-degrading signal augmentations to the input and capture the changes in embeddings and model logits. We guide the model towards salient features and away from non-salient features using a contrastive triplet loss. Additionally, we perform a sanity check on the logit distributions to ensure that the model outputs match the saliency-based augmentations. We demonstrate a boost in classification performance across both open- and closed-set scenarios against SOTA saliency-based methods, showing SAGE's effectiveness across various backbones, and include experiments to suggest its wide generalization across tasks. --- ### An Evaluation Framework for Network IDS/IPS Datasets: Leveraging MITRE ATT&CK and Industry Relevance Metrics Authors: Adrita Rahman Tori, Khondokar Fida Hasan Published: 2025-11-16 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2511.12743v1 The performance of Machine Learning (ML) and Deep Learning (DL)-based Intrusion Detection and Prevention Systems (IDS/IPS) is critically dependent on the relevance and quality of the datasets used for training and evaluation. However, current AI model evaluation practices for developing IDS/IPS focus predominantly on accuracy metrics, often overlooking whether datasets represent industry-specific threats. To address this gap, we introduce a novel multi-dimensional framework that integrates the MITRE ATT&CK knowledge base for threat intelligence and employs five complementary metrics that together provide a comprehensive assessment of dataset suitability. Methodologically, this framework combines threat intelligence, natural language processing, and quantitative analysis to assess the suitability of datasets for specific industry contexts. Applying this framework to nine publicly available IDS/IPS datasets reveals significant gaps in threat coverage, particularly in the healthcare, energy, and financial sectors. In particular, recent datasets (e.g., CIC-IoMT, CIC-UNSW-NB15) align better with sector-specific threats, whereas others, like CICIoV-24, underperform despite their recency. Our findings provide a standardized, interpretable approach for selecting datasets aligned with sector-specific operational requirements, ultimately enhancing the real-world effectiveness of AI-driven IDS/IPS deployments. The efficiency and practicality of the framework are validated through deployment in a real-world case study, underscoring its capacity to inform dataset selection and enhance the effectiveness of AI-driven IDS/IPS in operational environments. --- ### Stabilizing Self-Consuming Diffusion Models with Latent Space Filtering Authors: Zhongteng Cai, Yaxuan Wang, Yang Liu, Xueru Zhang Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12742v1 As synthetic data proliferates across the Internet, it is often reused to train successive generations of generative models. This creates a ``self-consuming loop" that can lead to training instability or \textit{model collapse}. Common strategies to address the issue -- such as accumulating historical training data or injecting fresh real data -- either increase computational cost or require expensive human annotation. In this paper, we empirically analyze the latent space dynamics of self-consuming diffusion models and observe that the low-dimensional structure of latent representations extracted from synthetic data degrade over generations. Based on this insight, we propose \textit{Latent Space Filtering} (LSF), a novel approach that mitigates model collapse by filtering out less realistic synthetic data from mixed datasets. Theoretically, we present a framework that connects latent space degradation to empirical observations. Experimentally, we show that LSF consistently outperforms existing baselines across multiple real-world datasets, effectively mitigating model collapse without increasing training cost or relying on human annotation. --- ### Deep Imbalanced Multi-Target Regression: 3D Point Cloud Voxel Content Estimation in Simulated Forests Authors: Amirhossein Hassanzadeh, Bartosz Krawczyk, Michael Saunders, Rob Wible, Keith Krause, Dimah Dera, Jan van Aardt Published: 2025-11-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.12740v1 Voxelization is an effective approach to reduce the computational cost of processing Light Detection and Ranging (LiDAR) data, yet it results in a loss of fine-scale structural information. This study explores whether low-level voxel content information, specifically target occupancy percentage within a voxel, can be inferred from high-level voxelized LiDAR point cloud data collected from Digital Imaging and remote Sensing Image Generation (DIRSIG) software. In our study, the targets include bark, leaf, soil, and miscellaneous materials. We propose a multi-target regression approach in the context of imbalanced learning using Kernel Point Convolutions (KPConv). Our research leverages cost-sensitive learning to address class imbalance called density-based relevance (DBR). We employ weighted Mean Saquared Erorr (MSE), Focal Regression (FocalR), and regularization to improve the optimization of KPConv. This study performs a sensitivity analysis on the voxel size (0.25 - 2 meters) to evaluate the effect of various grid representations in capturing the nuances of the forest. This sensitivity analysis reveals that larger voxel sizes (e.g., 2 meters) result in lower errors due to reduced variability, while smaller voxel sizes (e.g., 0.25 or 0.5 meter) exhibit higher errors, particularly within the canopy, where variability is greatest. For bark and leaf targets, error values at smaller voxel size datasets (0.25 and 0.5 meter) were significantly higher than those in larger voxel size datasets (2 meters), highlighting the difficulty in accurately estimating within-canopy voxel content at fine resolutions. This suggests that the choice of voxel size is application-dependent. Our work fills the gap in deep imbalance learning models for multi-target regression and simulated datasets for 3D LiDAR point clouds of forests. --- ### Direct Visual Grounding by Directing Attention of Visual Tokens Authors: Parsa Esmaeilkhani, Longin Jan Latecki Published: 2025-11-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.12738v2 Vision Language Models (VLMs) mix visual tokens and text tokens. A puzzling issue is the fact that visual tokens most related to the query receive little to no attention in the final layers of the LLM module of VLMs from the answer tokens, where all tokens are treated equally, in particular, visual and language tokens in the LLM attention layers. This fact may result in wrong answers to visual questions, as our experimental results confirm. It appears that the standard next-token prediction (NTP) loss provides an insufficient signal for directing attention to visual tokens. We hypothesize that a more direct supervision of the attention of visual tokens to corresponding language tokens in the LLM module of VLMs will lead to improved performance on visual tasks. To demonstrate that this is indeed the case, we propose a novel loss function that directly supervises the attention of visual tokens. It directly grounds the answer language tokens in images by directing their attention to the relevant visual tokens. This is achieved by aligning the attention distribution of visual tokens to ground truth attention maps with KL divergence. The ground truth attention maps are obtained from task geometry in synthetic cases or from standard grounding annotations (e.g., bounding boxes or point annotations) in real images, and are used inside the LLM for attention supervision without requiring new labels. The obtained KL attention loss (KLAL) when combined with NTP encourages VLMs to attend to relevant visual tokens while generating answer tokens. This results in notable improvements across geometric tasks, pointing, and referring expression comprehension on both synthetic and real-world data, as demonstrated by our experiments. We also introduce a new dataset to evaluate the line tracing abilities of VLMs. Surprisingly, even commercial VLMs do not perform well on this task. --- ### Backdoor Attacks on Open Vocabulary Object Detectors via Multi-Modal Prompt Tuning Authors: Ankita Raj, Chetan Arora Published: 2025-11-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.12735v2 Open-vocabulary object detectors (OVODs) unify vision and language to detect arbitrary object categories based on text prompts, enabling strong zero-shot generalization to novel concepts. As these models gain traction in high-stakes applications such as robotics, autonomous driving, and surveillance, understanding their security risks becomes crucial. In this work, we conduct the first study of backdoor attacks on OVODs and reveal a new attack surface introduced by prompt tuning. We propose TrAP (Trigger-Aware Prompt tuning), a multi-modal backdoor injection strategy that jointly optimizes prompt parameters in both image and text modalities along with visual triggers. TrAP enables the attacker to implant malicious behavior using lightweight, learnable prompt tokens without retraining the base model weights, thus preserving generalization while embedding a hidden backdoor. We adopt a curriculum-based training strategy that progressively shrinks the trigger size, enabling effective backdoor activation using small trigger patches at inference. Experiments across multiple datasets show that TrAP achieves high attack success rates for both object misclassification and object disappearance attacks, while also improving clean image performance on downstream datasets compared to the zero-shot setting. Code: https://github.com/rajankita/TrAP --- ### Improving the Generalisation of Learned Reconstruction Frameworks Authors: Emilien Valat, Ozan Öktem Published: 2025-11-16 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2511.12730v1 Ensuring proper generalization is a critical challenge in applying data-driven methods for solving inverse problems in imaging, as neural networks reconstructing an image must perform well across varied datasets and acquisition geometries. In X-ray Computed Tomography (CT), convolutional neural networks (CNNs) are widely used to filter the projection data but are ill-suited for this task as they apply grid-based convolutions to the sinogram, which inherently lies on a line manifold, not a regular grid. The CNNs, unaware of the geometry, are implicitly tied to it and require an excessive amount of parameters as they must infer the relations between measurements from the data rather than from prior information. The contribution of this paper is twofold. First, we introduce a graph data structure to represent CT acquisition geometries and tomographic data, providing a detailed explanation of the graph's structure for circular, cone-beam geometries. Second, we propose GLM, a hybrid neural network architecture that leverages both graph and grid convolutions to process tomographic data. We demonstrate that GLM outperforms CNNs when performance is quantified in terms of structural similarity and peak signal-to-noise ratio, despite the fact that GLM uses only a fraction of the trainable parameters. Compared to CNNs, GLM also requires significantly less training time and memory, and its memory requirements scale better. Crucially, GLM demonstrates robust generalization to unseen variations in the acquisition geometry, like when training only on fully sampled CT data and then testing on sparse-view CT data. --- ### On the Brittleness of LLMs: A Journey around Set Membership Authors: Lea Hergert, Gábor Berend, Mario Szegedy, Gyorgy Turan, Márk Jelasity Published: 2025-11-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.12728v1 Large language models (LLMs) achieve superhuman performance on complex reasoning tasks, yet often fail on much simpler problems, raising concerns about their reliability and interpretability. We investigate this paradox through a focused study with two key design features: simplicity, to expose basic failure modes, and scale, to enable comprehensive controlled experiments. We focus on set membership queries -- among the most fundamental forms of reasoning -- using tasks like ``Is apple an element of the set \{pear, plum, apple, raspberry\}?''. We conduct a systematic empirical evaluation across prompt phrasing, semantic structure, element ordering, and model choice. Our large-scale analysis reveals that LLM performance on this elementary task is consistently brittle, and unpredictable across all dimensions, suggesting that the models' ``understanding'' of the set concept is fragmented and convoluted at best. Our work demonstrates that the large-scale experiments enabled by the simplicity of the problem allow us to map and analyze the failure modes comprehensively, making this approach a valuable methodology for LLM evaluation in general. --- ### Convolutional Model Trees Authors: William Ward Armstrong, Hongyi Li, Jun Xu Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12725v2 A method for creating a forest of model trees to fit samples of a function defined on images is described in several steps: down-sampling the images, determining a tree's hyperplanes, applying convolutions to the hyperplanes to handle small distortions of training images, and creating forests of model trees to increase accuracy and achieve a smooth fit. A 1-to-1 correspondence among pixels of images, coefficients of hyperplanes and coefficients of leaf functions offers the possibility of dealing with larger distortions such as arbitrary rotations or changes of perspective. A theoretical method for smoothing forest outputs to produce a continuously differentiable approximation is described. Within that framework, a training procedure is proved to converge. --- ### LAYA: Layer-wise Attention Aggregation for Interpretable Depth-Aware Neural Networks Authors: Gennaro Vessio Published: 2025-11-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.12723v2 Deep neural networks typically rely on the representation produced by their final hidden layer to make predictions, implicitly assuming that this single vector fully captures the semantics encoded across all preceding transformations. However, intermediate layers contain rich and complementary information---ranging from low-level patterns to high-level abstractions---that is often discarded when the decision head depends solely on the last representation. This paper revisits the role of the output layer and introduces LAYA (Layer-wise Attention Aggregator), a novel output head that dynamically aggregates internal representations through attention. Instead of projecting only the deepest embedding, LAYA learns input-conditioned attention weights over layer-wise features, yielding an interpretable and architecture-agnostic mechanism for synthesizing predictions. Beyond improving feature aggregation, the learned attention coefficients provide intrinsic layer-attribution scores that explicitly quantify the contribution of each representation to the final decision, without requiring external post-hoc explanation methods. Experiments on image classification datasets show that LAYA achieves competitive predictive performance while producing meaningful depth-aware explanations. Furthermore, quantitative and qualitative interpretability analyses demonstrate that LAYA's attention scores closely reflect the actual contribution of individual layers, revealing structured, task-dependent patterns of depth utilization while providing intuitive explanations of how different abstraction levels contribute to each prediction. ---