# Research Papers — 2025-11 · bundle page 23 of 197 > Full-text markdown of papers #1101–1150 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/23 [← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/22.md) [Next bundle →](https://aipod.dev/papers/archive/2025-11/page/24.md) ### Can Multi-Modal LLMs Provide Live Step-by-Step Task Guidance? Authors: Apratim Bhattacharyya, Bicheng Xu, Sanjay Haresh, Reza Pourreza, Litian Liu, Sunny Panchal, Pulkit Madan, Leonid Sigal, Roland Memisevic Published: 2025-11-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.21998v2 Multi-modal Large Language Models (LLM) have advanced conversational abilities but struggle with providing live, interactive step-by-step guidance, a key capability for future AI assistants. Effective guidance requires not only delivering instructions but also detecting their successful execution, as well as identifying and alerting users to mistakes, all of which has to happen in real-time. This requires models that are not turn-based, but that can react asynchronously to a video stream, as well as video data showing users performing tasks including mistakes and their corrections. To this end, we introduce Qualcomm Interactive Cooking, a new benchmark and dataset built upon CaptainCook4D, which contains user mistakes during task execution. Our dataset and benchmark features densely annotated, timed instructions and feedback messages, specifically including mistake alerts precisely timestamped to their visual occurrence in the video. We evaluate state-of-the-art multi-modal LLMs on the Qualcomm Interactive Cooking benchmark and introduce LiveMamba, a streaming multi-modal LLM designed for interactive instructional guidance. This work provides the first dedicated benchmark and a strong baseline for developing and evaluating on live, situated coaching. --- ### Joint Estimation of Sea State and Vessel Parameters Using a Mass-Spring-Damper Equivalence Model Authors: Ranjeet K. Tiwari, Daniel Sgarioto, Peter Graham, Alexei Skvortsov, Sanjeev Arulampalam, Damith C. Ranasinghe Published: 2025-11-27 Categories: eess.SP, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2511.21997v2 Real-time sea state estimation is vital for applications like shipbuilding and maritime safety. Traditional methods rely on accurate wave-vessel transfer functions to estimate wave spectra from onboard sensors. In contrast, our approach jointly estimates sea state and vessel parameters without needing prior transfer function knowledge, which may be unavailable or variable. We model the wave-vessel system using pseudo mass-spring-dampers and develop a dynamic model for the system. This method allows for recursive modeling of wave excitation as a time-varying input, relaxing prior works' assumption of a constant input. We derive statistically consistent process noise covariance and implement a square root cubature Kalman filter for sensor data fusion. Further, we derive the Posterior Cramer-Rao lower bound to evaluate estimator performance. Extensive Monte Carlo simulations and data from a high-fidelity validated simulator confirm that the estimated wave spectrum matches methods assuming complete transfer function knowledge. --- ### A Safety and Security Framework for Real-World Agentic Systems Authors: Shaona Ghosh, Barnaby Simkin, Kyriacos Shiarlis, Soumili Nandi, Dan Zhao, Matthew Fiedler, Julia Bazinska, Nikki Pope, Roopa Prabhu, Daniel Rohrer, Michael Demoret, Bartley Richardson Published: 2025-11-27 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2511.21990v1 This paper introduces a dynamic and actionable framework for securing agentic AI systems in enterprise deployment. We contend that safety and security are not merely fixed attributes of individual models but also emergent properties arising from the dynamic interactions among models, orchestrators, tools, and data within their operating environments. We propose a new way of identification of novel agentic risks through the lens of user safety. Although, for traditional LLMs and agentic models in isolation, safety and security has a clear separation, through the lens of safety in agentic systems, they appear to be connected. Building on this foundation, we define an operational agentic risk taxonomy that unifies traditional safety and security concerns with novel, uniquely agentic risks, including tool misuse, cascading action chains, and unintended control amplification among others. At the core of our approach is a dynamic agentic safety and security framework that operationalizes contextual agentic risk management by using auxiliary AI models and agents, with human oversight, to assist in contextual risk discovery, evaluation, and mitigation. We further address one of the most challenging aspects of safety and security of agentic systems: risk discovery through sandboxed, AI-driven red teaming. We demonstrate the framework effectiveness through a detailed case study of NVIDIA flagship agentic research assistant, AI-Q Research Assistant, showcasing practical, end-to-end safety and security evaluations in complex, enterprise-grade agentic workflows. This risk discovery phase finds novel agentic risks that are then contextually mitigated. We also release the dataset from our case study, containing traces of over 10,000 realistic attack and defense executions of the agentic workflow to help advance research in agentic safety. --- ### CAT: A Metric-Driven Framework for Analyzing the Consistency-Accuracy Relation of LLMs under Controlled Input Variations Authors: Paulo Cavalin, Cassia Sanctos, Marcelo Grave, Claudio Pinhanez, Yago Primerano Published: 2025-11-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.23711v1 We introduce \textsc{CAT}, a framework designed to evaluate and visualize the \emph{interplay} of \emph{accuracy} and \emph{response consistency} of Large Language Models (LLMs) under controllable input variations, using multiple-choice (MC) benchmarks as a case study. Current evaluation practices primarily focus on model capabilities such as accuracy or benchmark scores and, more recently, measuring consistency is being considered an essential property for deploying LLMs in high-stake, real-world applications. We argue in this paper that although both dimensions should still be evaluated independently, their inter-dependency also need to be considered for a more nuanced evaluation of LLMs. At the core of \textsc{CAT} are the \emph{Consistency-Accuracy Relation (CAR)} curves, which visualize how model accuracy varies with increasing consistency requirements, as defined by the \emph{Minimum-Consistency Accuracy (MCA)} metric. We further propose the \emph{Consistency-Oriented Robustness Estimate (CORE)} index, a global metric that combines the area and shape of the CAR curve to quantify the trade-off between accuracy and consistency. We present a practical demonstration of our framework across a diverse set of generalist and domain-specific LLMs, evaluated on multiple MC benchmarks. We also outline how \textsc{CAT} can be extended beyond MC tasks to support long-form, open-ended evaluations through adaptable scoring functions. --- ### GPU-Virt-Bench: A Comprehensive Benchmarking Framework for Software-Based GPU Virtualization Systems Authors: Jithin VG, Ditto PS Published: 2025-11-26 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2512.22125v1 The proliferation of GPU-accelerated workloads, particularly in artificial intelligence and large language model (LLM) inference, has created unprecedented demand for efficient GPU resource sharing in cloud and container environments. While NVIDIA's Multi-Instance GPU (MIG) technology provides hardware-level isolation, its availability is limited to high-end datacenter GPUs. Software-based virtualization solutions such as HAMi-core and BUD-FCSP offer alternatives for broader GPU families but lack standardized evaluation methodologies. We present GPU-Virt-Bench, a comprehensive benchmarking framework that evaluates GPU virtualization systems across 56 performance metrics organized into 10 categories. Our framework measures overhead, isolation quality, LLM-specific performance, memory bandwidth, cache behavior, PCIe throughput, multi-GPU communication, scheduling efficiency, memory fragmentation, and error recovery. GPU-Virt-Bench enables systematic comparison between software virtualization approaches and ideal MIG behavior, providing actionable insights for practitioners deploying GPU resources in multi-tenant environments. We demonstrate the framework's utility through evaluation of HAMi-core, BUD-FCSP, and simulated MIG baselines, revealing performance characteristics critical for production deployment decisions. --- ### MegaRAG: Multimodal Knowledge Graph-Based Retrieval Augmented Generation Authors: Chi-Hsiang Hsiao, Yi-Cheng Wang, Tzung-Sheng Lin, Yi-Ren Yeh, Chu-Song Chen Published: 2025-11-26 Categories: cs.AI, cs.CL, cs.CV, cs.IR arXiv: https://arxiv.org/abs/2512.20626v2 Retrieval-augmented generation (RAG) enables large language models (LLMs) to dynamically access external information, which is powerful for answering questions over previously unseen documents. Nonetheless, they struggle with high-level conceptual understanding and holistic comprehension due to limited context windows, which constrain their ability to perform deep reasoning over long-form, domain-specific content such as full-length books. To solve this problem, knowledge graphs (KGs) have been leveraged to provide entity-centric structure and hierarchical summaries, offering more structured support for reasoning. However, existing KG-based RAG solutions remain restricted to text-only inputs and fail to leverage the complementary insights provided by other modalities such as vision. On the other hand, reasoning from visual documents requires textual, visual, and spatial cues into structured, hierarchical concepts. To address this issue, we introduce a multimodal knowledge graph-based RAG that enables cross-modal reasoning for better content understanding. Our method incorporates visual cues into the construction of knowledge graphs, the retrieval phase, and the answer generation process. Experimental results across both global and fine-grained question answering tasks show that our approach consistently outperforms existing RAG-based approaches on both textual and multimodal corpora. --- ### Graph-O1 : Monte Carlo Tree Search with Reinforcement Learning for Text-Attributed Graph Reasoning Authors: Lihui Liu Published: 2025-11-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.17912v1 ChatGPT said: Text-attributed graphs, where nodes and edges contain rich textual information, are widely used across diverse domains. A central challenge in this setting is question answering, which requires jointly leveraging unstructured text and the structured relational signals within the graph. Although Large Language Models (LLMs) have made significant advances in natural language understanding, their direct use for reasoning over text-attributed graphs remains limited. Retrieval-augmented generation methods that operate purely on text often treat passages as isolated units, ignoring the interconnected structure of the graph. Conversely, graph-based RAG methods that serialize large subgraphs into long textual sequences quickly become infeasible due to LLM context-length constraints, resulting in fragmented reasoning and degraded accuracy. To overcome these limitations, we introduce Graph-O1, an agentic GraphRAG framework that enables LLMs to conduct stepwise, interactive reasoning over graphs. Our approach integrates Monte Carlo Tree Search (MCTS) with end-to-end reinforcement learning, allowing the model to selectively explore and retrieve only the most informative subgraph components. The reasoning procedure is framed as a multi-turn interaction between the agent and the graph environment, and the agent is trained through a unified reward mechanism. Extensive experiments across multiple LLM backbones demonstrate that Graph-O1 consistently surpasses state-of-the-art baselines, producing answers that are more accurate, reliable, and interpretable. --- ### Towards Reasoning-Preserving Unlearning in Multimodal Large Language Models Authors: Hongji Li, Junchi yao, Manjiang Yu, Priyanka Singh, Xue Li, Di Wang, Lijie Hu Published: 2025-11-26 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.17911v1 Machine unlearning aims to erase requested data from trained models without full retraining. For Reasoning Multimodal Large Language Models (RMLLMs), this is uniquely challenging: intermediate chain-of-thought steps can still leak sensitive information even when final answers are forgotten, and overly aggressive interventions easily damage general reasoning ability. Yet no benchmark jointly evaluates how well unlearning methods suppress reasoning-level leakage while preserving reasoning competence. We address this gap with RMLLMU-Bench, the first benchmark for RMLLM unlearning that extends standard forgetting metrics with dedicated measures of reasoning leakage and reasoning retention. A systematic evaluation on RMLLMU-Bench reveals that existing unlearning methods for MLLMs and Large (Language) Reasoning Models (LRMs) either leave substantial leakage in the reasoning process or severely degrade reasoning performance. To address these gaps, we propose R-MUSE (Reasoning-preserving MLLM Unlearning via Subspace guidance and Adaptive Steering), a training-free and inference-time intervention framework that steers internal representations to forget both answers and reasoning traces while explicitly preserving general reasoning. Experiments on RMLLMU-Bench demonstrate that R-MUSE achieves a substantially better balance between effective forgetting and reasoning retention. --- ### Efficient Multi-Adapter LLM Serving via Cross-Model KV-Cache Reuse with Activated LoRA Authors: Allison Li, Kristjan Greenewald, Thomas Parnell, Navid Azizan Published: 2025-11-26 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.17910v1 Modern large language model (LLM) systems increasingly rely on multi-turn pipelines that are composed of multiple task-specific adapters, yet existing serving frameworks remain inefficient, incurring substantial recomputation overhead when switching between adapters. We present the first LLM serving engine that supports cross-model prefix cache reuse between base and adapted models via Activated LoRA (aLoRA), enabling efficient and fine-grained adapter switching during inference. Our design extends the vLLM framework by introducing base-aligned block hashing and activation-aware masking within the model execution path, permitting cache reuse across models while preserving compatibility with existing serving engine optimizations. Integrated into a production-grade inference stack, this approach supports dynamic adapter activation without excessive key-value tensor recomputation. Evaluation across representative multi-turn, multi-adapter pipelines demonstrates up to 58x end-to-end latency reduction and over 100x time-to-first-token improvement relative to standard LoRA baselines, with benefits that scale with model size and sequence length and manifest across all stages of the request lifecycle. This work bridges parameter-efficient model adaptation with high-performance serving, providing the first complete realization of cross-model KV-cache reuse in modern LLM inference engines. --- ### Reinforcement Learning for Latent-Space Thinking in LLMs Authors: Enes Özeren, Matthias Aßenmacher Published: 2025-11-26 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2512.11816v1 Chain-of-Thought (CoT) reasoning typically utilizes the discrete language space for thinking, which is inherently inefficient, as many generated tokens only enforce linguistic rules that are not required for reasoning. To bypass this, latent-space thinking allows models to think using the continuous embedding space. While existing methods for training those models show domain-specific gains, they fail to maintain performance in complex tasks, such as mathematical reasoning. We experimentally demonstrate that the Coconut approach, a form of supervised fine-tuning for latent-space thinking, is highly sensitive to design choices and exhibits several inherent limitations. To address these issues, we investigate reinforcement learning (RL) techniques -- an underexplored direction in latent-space thinking -- including GRPO and design a novel Latent RL method for directly optimizing the latent thinking steps. Our experimental results reveal that these RL-trained models still lag behind traditional language-space CoT models in the mathematical reasoning domain. We make our codebase publicly available. --- ### Totalitarian Technics: The Hidden Cost of AI Scribes in Healthcare Authors: Hugh Brosnahan Published: 2025-11-26 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2512.11814v1 Artificial intelligence (AI) scribes, systems that record and summarise patient-clinician interactions, are promoted as solutions to administrative overload. This paper argues that their significance lies not in efficiency gains but in how they reshape medical attention itself. Offering a conceptual analysis, it situates AI scribes within a broader philosophical lineage concerned with the externalisation of human thought and skill. Drawing on Iain McGilchrist's hemisphere theory and Lewis Mumford's philosophy of technics, the paper examines how technology embodies and amplifies a particular mode of attention. AI scribes, it contends, exemplify the dominance of a left-hemispheric, calculative mindset that privileges the measurable and procedural over the intuitive and relational. As this mode of attention becomes further embedded in medical practice, it risks narrowing the field of care, eroding clinical expertise, and reducing physicians to operators within an increasingly mechanised system. --- ### Quantum Circuit Reasoning Models: A Variational Framework for Differentiable Logical Inference Authors: Andrew Kiruluta Published: 2025-11-26 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2512.07871v1 This report introduces a novel class of reasoning architectures, termed Quantum Circuit Reasoning Models (QCRM), which extend the concept of Variational Quantum Circuits (VQC) from energy minimization and classification tasks to structured logical inference and reasoning. We posit that fundamental quantum mechanical operations, superposition, entanglement, interference, and measurement, naturally map to essential reasoning primitives such as hypothesis branching, constraint propagation, consistency enforcement, and decision making. The resulting framework combines quantum-inspired computation with differentiable optimization, enabling reasoning to emerge as a process of amplitude evolution and interference-driven selection of self-consistent states. We develop the mathematical foundation of QCRM, define its parameterized circuit architecture, and show how logical rules can be encoded as unitary transformations over proposition-qubit states. We further formalize a training objective grounded in classical gradient descent over circuit parameters and discuss simulation-based implementations on classical hardware. Finally, we propose the Quantum Reasoning Layer (QRL) as a differentiable hybrid component for composable reasoning models applicable to scientific, biomedical, and chemical inference domains. --- ### Manifolds and Modules: How Function Develops in a Neural Foundation Model Authors: Johannes Bertram, Luciano Dyballa, T. Anderson Keller, Savik Kinger, Steven W. Zucker Published: 2025-11-26 Categories: q-bio.NC, cs.AI arXiv: https://arxiv.org/abs/2512.07869v1 Foundation models have shown remarkable success in fitting biological visual systems; however, their black-box nature inherently limits their utility for understanding brain function. Here, we peek inside a SOTA foundation model of neural activity (Wang et al., 2025) as a physiologist might, characterizing each 'neuron' based on its temporal response properties to parametric stimuli. We analyze how different stimuli are represented in neural activity space by building decoding manifolds, and we analyze how different neurons are represented in stimulus-response space by building neural encoding manifolds. We find that the different processing stages of the model (i.e., the feedforward encoder, recurrent, and readout modules) each exhibit qualitatively different representational structures in these manifolds. The recurrent module shows a jump in capabilities over the encoder module by 'pushing apart' the representations of different temporal stimulus patterns; while the readout module achieves biological fidelity by using numerous specialized feature maps rather than biologically plausible mechanisms. Overall, we present this work as a study of the inner workings of a prominent neural foundation model, gaining insights into the biological relevance of its internals through the novel analysis of its neurons' joint temporal response patterns. --- ### Bayesian Optimization for Function-Valued Responses under Min-Max Criteria Authors: Pouya Ahadi, Reza Marzban, Ali Adibi, Kamran Paynabar Published: 2025-11-26 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2512.07868v2 Bayesian optimization is widely used for optimizing expensive black box functions, but most existing approaches focus on scalar responses. In many scientific and engineering settings the response is functional, varying smoothly over an index such as time or wavelength, which makes classical formulations inadequate. Existing methods often minimize integrated error, which captures average performance but neglects worst case deviations. To address this limitation we propose min-max Functional Bayesian Optimization (MM-FBO), a framework that directly minimizes the maximum error across the functional domain. Functional responses are represented using functional principal component analysis, and Gaussian process surrogates are constructed for the principal component scores. Building on this representation, MM-FBO introduces an integrated uncertainty acquisition function that balances exploitation of worst case expected error with exploration across the functional domain. We provide two theoretical guarantees: a discretization bound for the worst case objective, and a consistency result showing that as the surrogate becomes accurate and uncertainty vanishes, the acquisition converges to the true min-max objective. We validate the method through experiments on synthetic benchmarks and physics inspired case studies involving electromagnetic scattering by metaphotonic devices and vapor phase infiltration. Results show that MM-FBO consistently outperforms existing baselines and highlights the importance of explicitly modeling functional uncertainty in Bayesian optimization. --- ### LLM-Generated Counterfactual Stress Scenarios for Portfolio Risk Simulation via Hybrid Prompt-RAG Pipeline Authors: Masoud Soleimani Published: 2025-11-26 Categories: q-fin.RM, cs.AI, econ.EM arXiv: https://arxiv.org/abs/2512.07867v1 We develop a transparent and fully auditable LLM-based pipeline for macro-financial stress testing, combining structured prompting with optional retrieval of country fundamentals and news. The system generates machine-readable macroeconomic scenarios for the G7, which cover GDP growth, inflation, and policy rates, and are translated into portfolio losses through a factor-based mapping that enables Value-at-Risk and Expected Shortfall assessment relative to classical econometric baselines. Across models, countries, and retrieval settings, the LLMs produce coherent and country-specific stress narratives, yielding stable tail-risk amplification with limited sensitivity to retrieval choices. Comprehensive plausibility checks, scenario diagnostics, and ANOVA-based variance decomposition show that risk variation is driven primarily by portfolio composition and prompt design rather than by the retrieval mechanism. The pipeline incorporates snapshotting, deterministic modes, and hash-verified artifacts to ensure reproducibility and auditability. Overall, the results demonstrate that LLM-generated macro scenarios, when paired with transparent structure and rigorous validation, can provide a scalable and interpretable complement to traditional stress-testing frameworks. --- ### Command & Control (C2) Traffic Detection Via Algorithm Generated Domain (Dga) Classification Using Deep Learning And Natural Language Processing Authors: Maria Milena Araujo Felix Published: 2025-11-26 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2512.07866v1 The sophistication of modern malware, specifically regarding communication with Command and Control (C2) servers, has rendered static blacklist-based defenses obsolete. The use of Domain Generation Algorithms (DGA) allows attackers to generate thousands of dynamic addresses daily, hindering blocking by traditional firewalls. This paper aims to propose and evaluate a method for detecting DGA domains using Deep Learning and Natural Language Processing (NLP) techniques. The methodology consisted of collecting a hybrid database containing 50,000 legitimate and 50,000 malicious domains, followed by the extraction of lexical features and the training of a Recurrent Neural Network (LSTM). Results demonstrated that while statistical entropy analysis is effective for simple DGAs, the Neural Network approach presents superiority in detecting complex patterns, reaching 97.2% accuracy and reducing the false positive rate in ambiguous lawful traffic scenarios. --- ### Using Text-Based Life Trajectories from Swedish Register Data to Predict Residential Mobility with Pretrained Transformers Authors: Philipp Stark, Alexandros Sopasakis, Ola Hall, Markus Grillitsch Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07865v1 We transform large-scale Swedish register data into textual life trajectories to address two long-standing challenges in data analysis: high cardinality of categorical variables and inconsistencies in coding schemes over time. Leveraging this uniquely comprehensive population register, we convert register data from 6.9 million individuals (2001-2013) into semantically rich texts and predict individuals' residential mobility in later years (2013-2017). These life trajectories combine demographic information with annual changes in residence, work, education, income, and family circumstances, allowing us to assess how effectively such sequences support longitudinal prediction. We compare multiple NLP architectures (including LSTM, DistilBERT, BERT, and Qwen) and find that sequential and transformer-based models capture temporal and semantic structure more effectively than baseline models. The results show that textualized register data preserves meaningful information about individual pathways and supports complex, scalable modeling. Because few countries maintain longitudinal microdata with comparable coverage and precision, this dataset enables analyses and methodological tests that would be difficult or impossible elsewhere, offering a rigorous testbed for developing and evaluating new sequence-modeling approaches. Overall, our findings demonstrate that combining semantically rich register data with modern language models can substantially advance longitudinal analysis in social sciences. --- ### Pattern Recognition of Ozone-Depleting Substance Exports in Global Trade Data Authors: Muhammad Sukri Bin Ramli Published: 2025-11-26 Categories: cs.LG, econ.EM, econ.GN arXiv: https://arxiv.org/abs/2512.07864v1 New methods are needed to monitor environmental treaties, like the Montreal Protocol, by reviewing large, complex customs datasets. This paper introduces a framework using unsupervised machine learning to systematically detect suspicious trade patterns and highlight activities for review. Our methodology, applied to 100,000 trade records, combines several ML techniques. Unsupervised Clustering (K-Means) discovers natural trade archetypes based on shipment value and weight. Anomaly Detection (Isolation Forest and IQR) identifies rare "mega-trades" and shipments with commercially unusual price-per-kilogram values. This is supplemented by Heuristic Flagging to find tactics like vague shipment descriptions. These layers are combined into a priority score, which successfully identified 1,351 price outliers and 1,288 high-priority shipments for customs review. A key finding is that high-priority commodities show a different and more valuable value-to-weight ratio than general goods. This was validated using Explainable AI (SHAP), which confirmed vague descriptions and high value as the most significant risk predictors. The model's sensitivity was validated by its detection of a massive spike in "mega-trades" in early 2021, correlating directly with the real-world regulatory impact of the US AIM Act. This work presents a repeatable unsupervised learning pipeline to turn raw trade data into prioritized, usable intelligence for regulatory groups. --- ### SetAD: Semi-Supervised Anomaly Learning in Contextual Sets Authors: Jianling Gao, Chongyang Tao, Xuelian Lin, Junfeng Liu, Shuai Ma Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07863v1 Semi-supervised anomaly detection (AD) has shown great promise by effectively leveraging limited labeled data. However, existing methods are typically structured around scoring individual points or simple pairs. Such {point- or pair-centric} view not only overlooks the contextual nature of anomalies, which are defined by their deviation from a collective group, but also fails to exploit the rich supervisory signals that can be generated from the combinatorial composition of sets. Consequently, such models struggle to exploit the high-order interactions within the data, which are critical for learning discriminative representations. To address these limitations, we propose SetAD, a novel framework that reframes semi-supervised AD as a Set-level Anomaly Detection task. SetAD employs an attention-based set encoder trained via a graded learning objective, where the model learns to quantify the degree of anomalousness within an entire set. This approach directly models the complex group-level interactions that define anomalies. Furthermore, to enhance robustness and score calibration, we propose a context-calibrated anomaly scoring mechanism, which assesses a point's anomaly score by aggregating its normalized deviations from peer behavior across multiple, diverse contextual sets. Extensive experiments on 10 real-world datasets demonstrate that SetAD significantly outperforms state-of-the-art models. Notably, we show that our model's performance consistently improves with increasing set size, providing strong empirical support for the set-based formulation of anomaly detection. --- ### Integrating LSTM Networks with Neural Levy Processes for Financial Forecasting Authors: Mohammed Alruqimi, Luca Di Persio Published: 2025-11-26 Categories: q-fin.ST, cs.LG arXiv: https://arxiv.org/abs/2512.07860v1 This paper investigates an optimal integration of deep learning with financial models for robust asset price forecasting. Specifically, we developed a hybrid framework combining a Long Short-Term Memory (LSTM) network with the Merton-Lévy jump-diffusion model. To optimise this framework, we employed the Grey Wolf Optimizer (GWO) for the LSTM hyperparameter tuning, and we explored three calibration methods for the Merton-Levy model parameters: Artificial Neural Networks (ANNs), the Marine Predators Algorithm (MPA), and the PyTorch-based TorchSDE library. To evaluate the predictive performance of our hybrid model, we compared it against several benchmark models, including a standard LSTM and an LSTM combined with the Fractional Heston model. This evaluation used three real-world financial datasets: Brent oil prices, the STOXX 600 index, and the IT40 index. Performance was assessed using standard metrics, including Mean Squared Error (MSE), Mean Absolute Error(MAE), Mean Squared Percentage Error (MSPE), and the coefficient of determination (R2). Our experimental results demonstrate that the hybrid model, combining a GWO-optimized LSTM network with the Levy-Merton Jump-Diffusion model calibrated using an ANN, outperformed the base LSTM model and all other models developed in this study. --- ### FAIM: Frequency-Aware Interactive Mamba for Time Series Classification Authors: Da Zhang, Bingyu Li, Zhiyuan Zhao, Yanhan Zhang, Junyu Gao, Feiping Nie, Xuelong Li Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07858v1 Time series classification (TSC) is crucial in numerous real-world applications, such as environmental monitoring, medical diagnosis, and posture recognition. TSC tasks require models to effectively capture discriminative information for accurate class identification. Although deep learning architectures excel at capturing temporal dependencies, they often suffer from high computational cost, sensitivity to noise perturbations, and susceptibility to overfitting on small-scale datasets. To address these challenges, we propose FAIM, a lightweight Frequency-Aware Interactive Mamba model. Specifically, we introduce an Adaptive Filtering Block (AFB) that leverages Fourier Transform to extract frequency-domain features from time series data. The AFB incorporates learnable adaptive thresholds to dynamically suppress noise and employs element-wise coupling of global and local semantic adaptive filtering, enabling in-depth modeling of the synergy among different frequency components. Furthermore, we design an Interactive Mamba Block (IMB) to facilitate efficient multi-granularity information interaction, balancing the extraction of fine-grained discriminative features and comprehensive global contextual information, thereby endowing FAIM with powerful and expressive representations for TSC tasks. Additionally, we incorporate a self-supervised pre-training mechanism to enhance FAIM's understanding of complex temporal patterns and improve its robustness across various domains and high-noise scenarios. Extensive experiments on multiple benchmarks demonstrate that FAIM consistently outperforms existing state-of-the-art (SOTA) methods, achieving a superior trade-off between accuracy and efficiency and exhibits outstanding performance. --- ### SA$^{2}$GFM: Enhancing Robust Graph Foundation Models with Structure-Aware Semantic Augmentation Authors: Junhua Shi, Qingyun Sun, Haonan Yuan, Xingcheng Fu Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07857v3 We present Graph Foundation Models (GFMs) which have made significant progress in various tasks, but their robustness against domain noise, structural perturbations, and adversarial attacks remains underexplored. A key limitation is the insufficient modeling of hierarchical structural semantics, which are crucial for generalization. In this paper, we propose SA$^{2}$GFM, a robust GFM framework that improves domain-adaptive representations through Structure-Aware Semantic Augmentation. First, we encode hierarchical structural priors by transforming entropy-based encoding trees into structure-aware textual prompts for feature augmentation. The enhanced inputs are processed by a self-supervised Information Bottleneck mechanism that distills robust, transferable representations via structure-guided compression. To address negative transfer in cross-domain adaptation, we introduce an expert adaptive routing mechanism, combining a mixture-of-experts architecture with a null expert design. For efficient downstream adaptation, we propose a fine-tuning module that optimizes hierarchical structures through joint intra- and inter-community structure learning. Extensive experiments demonstrate that SA$^{2}$GFM outperforms 9 state-of-the-art baselines in terms of effectiveness and robustness against random noise and adversarial perturbations for node and graph classification. --- ### Medical Test-free Disease Detection Based on Big Data Authors: Haokun Zhao, Yingzhe Bai, Qingyang Xu, Lixin Zhou, Jianxin Chen, Jicong Fan Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07856v1 Accurate disease detection is of paramount importance for effective medical treatment and patient care. However, the process of disease detection is often associated with extensive medical testing and considerable costs, making it impractical to perform all possible medical tests on a patient to diagnose or predict hundreds or thousands of diseases. In this work, we propose Collaborative Learning for Disease Detection (CLDD), a novel graph-based deep learning model that formulates disease detection as a collaborative learning task by exploiting associations among diseases and similarities among patients adaptively. CLDD integrates patient-disease interactions and demographic features from electronic health records to detect hundreds or thousands of diseases for every patient, with little to no reliance on the corresponding medical tests. Extensive experiments on a processed version of the MIMIC-IV dataset comprising 61,191 patients and 2,000 diseases demonstrate that CLDD consistently outperforms representative baselines across multiple metrics, achieving a 6.33\% improvement in recall and 7.63\% improvement in precision. Furthermore, case studies on individual patients illustrate that CLDD can successfully recover masked diseases within its top-ranked predictions, demonstrating both interpretability and reliability in disease prediction. By reducing diagnostic costs and improving accessibility, CLDD holds promise for large-scale disease screening and social health security. --- ### LAPA: Log-Domain Prediction-Driven Dynamic Sparsity Accelerator for Transformer Model Authors: Huizheng Wang, Hongbin Wang, Shaojun Wei, Yang Hu, Shouyi Yin Published: 2025-11-26 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2512.07855v1 Attention-based Transformers have revolutionized natural language processing (NLP) and shown strong performance in computer vision (CV) tasks. However, as the input sequence varies, the computational bottlenecks in Transformer models exhibit dynamic behavior across stages, which calls for a cross-stage sparse acceleration strategy. Unfortunately, most existing sparse Transformer approaches are single-stage based, and their sparsity prediction mechanisms lead to significant power overhead when applied across multiple stages. To this end, this paper proposes a log-domain attention prediction algorithm-architecture co-design, named LAPA. First, an asymmetric leading one computing (ALOC) scheme is designed to eliminate expensive multiplications. Next, a mixed-precision multi-round shifting accumulation (MRSA) mechanism is further proposed to mitigate the accumulation overhead. A data-feature dependent filter (DDF) strategy is designed to work in concert with the MRSA process. Finally, an elaborate accelerator is designed to translate the theoretical enhancement into practical hardware improvement. Experimental results show that LAPA achieves 3.52x, 3.24x and 2.79x higher energy efficiency than the state-of-the-art (SOTA) works Spatten, Sanger and FACT, respectively. --- ### HieraMix: A Hierarchical MLP-Mixer for Large-Scale Traffic Forecasting Authors: Yongyao Wang, Xie Yu, Jingyuan Wang, Jiahao Ji, Chao Li Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07854v2 Traffic forecasting task is significant to modern urban management. Recently, there is growing attention on large-scale forecasting, as it better reflects the complexity of real-world traffic networks. However, existing models often exhibit quadratic computational complexity, making them impractical for large-scale real-world scenarios. In this paper, we propose a novel framework, Spatio-Temporal Hierarchical Mixer (HieraMix), which leverages an all-MLP architecture for efficient and effective large-scale traffic forecasting. HieraMix employs a hierarchical spatiotemporal mixing block to extract multi-resolution features through bottom-up aggregation and top-down propagation. Furthermore, an adaptive region mixer generates transformation matrices based on regional semantics, enabling our model to dynamically capture evolving spatiotemporal patterns for different regions. Extensive experiments conducted on four large-scale real-world datasets demonstrate that the proposed method not only achieves state-of-the-art performance but also exhibits competitive computational efficiency. --- ### GPU Memory Prediction for Multimodal Model Training Authors: Jinwoo Jeong, Minchul Kang, Younghun Go, Changyong Shin, Hyunho Lee, Junho Yoon, Gyeongsik Yang, Chuck Yoo Published: 2025-11-26 Categories: cs.LG, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2512.07853v1 As deep learning models in agentic AI systems grow in scale and complexity, GPU memory requirements increase and often exceed the available GPU memory capacity, so that out-of-memory (OoM) errors occur. It is well known that OoM interrupts the whole training itself and wastes substantial computational resources. Therefore, to prevent OoM, accurate prediction of GPU memory usage is essential. However, previous studies focus only on unimodal architectures and fail to generalize to multimodal models, even though the multimodal models are a common choice in agentic AI systems. To address this limitation, we propose a framework that predicts the peak GPU memory usage by analyzing the model architecture and training behavior of multimodal models. Specifically, the framework decomposes the multimodal model into its constituent layers and applies factorization to estimate the memory usage of each layer. Our evaluation shows that our framework achieves high prediction accuracy of ~8.7% average MAPE. --- ### SABER: Small Actions, Big Errors -- Safeguarding Mutating Steps in LLM Agents Authors: Alejandro Cuadron, Pengfei Yu, Yang Liu, Arpit Gupta Published: 2025-11-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.07850v1 Despite rapid progress in LLM agents, performance on long-horizon, tool-using tasks remains fragile. To better understand this fragility, we ask a simple question: \emph{do all actions contribute equally to failure?} Analyzing execution traces on $τ$-Bench (Airline/Retail) and SWE-Bench Verified, we decompose trajectories into \emph{mutating} (environment-changing) vs.\ non-mutating steps and formalize \emph{decisive deviations}, earliest action, level divergences that flip success to failure. A logistic regression reveals that each additional deviation in a mutating action reduces the odds of success by upto $92\%$ on Airline and upto $96\%$ on Retail for SoTA models. In contrast, deviations in non-mutating actions have little to no effect. Errors also grow with context length as agents drift from role and act on stale constraints. Motivated by these observations, we introduce \cm{}, a model-agnostic, gradient-free, test-time safeguard that (i) adds mutation-gated verification, (ii) injects \emph{Targeted Reflection} before mutating steps, and (iii) performs block-based context cleaning. \cm{} delivers consistent gains, e.g., Qwen3-Thinking: +28\% \emph{relative} on Airline, +11\% on Retail, and +7\% on SWE-Bench Verified; Claude: +9\%/+7\%. We further identify ceiling effects in $τ$-Bench, where annotation errors and underspecified tasks artificially cap model performance. To address this, we release $τ$-Bench Verified, which restores benchmark headroom through targeted revisions. Our results argue for action-level analysis, targeted safeguards, and reliable evaluations as prerequisites for robust multi-turn agents. --- ### AI Urban Scientist: Multi-Agent Collaborative Automation for Urban Research Authors: Tong Xia, Jiankun Zhang, Ruiwen You, Ao Xu, Linghao Zhang, Tengyao Tu, Jingzhi Wang, Jinghua Piao, Yunke Zhang, Fengli Xu, Yong Li Published: 2025-11-26 Categories: cs.CY, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2512.07849v2 Urban research aims to understand how cities operate and evolve as complex adaptive systems. With the rapid growth of urban data and analytical methodologies, the central challenge of the field has shifted from data availability to the integration of heterogeneous data into coherent, verifiable urban knowledge through multidisciplinary approaches. Recent advances in AI, particularly the emergence of large language models (LLMs), have enabled the development of AI scientists capable of autonomous reasoning, hypothesis generation, and data-driven experimentation, demonstrating substantial potential for autonomous urban research. However, most general-purpose AI systems remain misaligned with the domain-specific knowledge, methodological conventions, and inferential standards required in urban studies. Here, we introduce the AI Urban Scientist, a knowledge-driven multi-agent framework designed to support autonomous urban research. Grounded in hypotheses, peer-review feedback, datasets, and research methodologies distilled from large-scale prior studies, the system constructs structured domain knowledge that guides LLM-based agents to automatically generate hypotheses, identify and integrate multi-source urban datasets, conduct empirical analyses and simulations, and iteratively refine analytical methods. Through this process, the framework synthesizes new insights in urban science and accelerates the urban research lifecycle. --- ### RaX-Crash: A Resource Efficient and Explainable Small Model Pipeline with an Application to City Scale Injury Severity Prediction Authors: Di Zhu, Chen Xie, Ziwei Wang, Haoyun Zhang Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.07848v1 New York City reports over one hundred thousand motor vehicle collisions each year, creating substantial injury and public health burden. We present RaX-Crash, a resource efficient and explainable small model pipeline for structured injury severity prediction on the official NYC Motor Vehicle Collisions dataset. RaX-Crash integrates three linked tables with tens of millions of records, builds a unified feature schema in partitioned storage, and trains compact tree based ensembles (Random Forest and XGBoost) on engineered tabular features, which are compared against locally deployed small language models (SLMs) prompted with textual summaries. On a temporally held out test set, XGBoost and Random Forest achieve accuracies of 0.7828 and 0.7794, clearly outperforming SLMs (0.594 and 0.496); class imbalance analysis shows that simple class weighting improves fatal recall with modest accuracy trade offs, and SHAP attribution highlights human vulnerability factors, timing, and location as dominant drivers of predicted severity. Overall, RaX-Crash indicates that interpretable small model ensembles remain strong baselines for city scale injury analytics, while hybrid pipelines that pair tabular predictors with SLM generated narratives improve communication without sacrificing scalability. --- ### SEB-ChOA: An Improved Chimp Optimization Algorithm Using Spiral Exploitation Behavior Authors: Leren Qian, Mohammad Khishe, Yiqian Huang, Seyedali Mirjalili Published: 2025-11-26 Categories: cs.NE arXiv: https://arxiv.org/abs/2512.05981v1 The chimp optimization algorithm (ChOA) is a nature-inspired algorithm that imitates chimpanzees' individual intelligence and hunting behaviors. In this algorithm, the hunting process consists of four steps: driving, blocking, chasing, and attacking. Because of the novelty of ChOA, the steps of the hunting process have been modeled in a simple way, leading to slow and premature convergence similar to other iterative algorithms. This paper proposes six spiral functions and introduces two novel hybrid spiral functions (SEB-ChOA) to address these deficiencies. The performance of SEB-ChOA is evaluated on 23 standard benchmarks, 20 benchmarks of the IEEE CEC-2005 test suite, 10 cases from the IEEE CEC06-2019 test suite, and 12 constrained real-world engineering problems from IEEE CEC-2020. The SEB-ChOA variants are compared with three groups of optimization algorithms, including Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) as well-known optimizers; Slime Mould Algorithm (SMA), Marine Predators Algorithm (MPA), Ant Lion Optimization (ALO), and Henry Gas Solubility Optimization (HGSO) as recently developed optimizers; and jDE100 and DISHchain1e+12, the winners of the IEEE CEC06-2019 competition. Additional comparisons are made with EBOwithCMAR and CIPDE as strong secondary baselines. The SEB-ChOA methods achieve top-ranked results on nearly all benchmarks and show competitive performance compared to jDE100 and DISHchain1e+12. Statistical results indicate that SEB-ChOA outperforms PSO, GA, SMA, MPA, ALO, and HGSO while producing results comparable to those of jDE100 and DISHchain1e+12. --- ### Accelerating Materials Discovery: Learning a Universal Representation of Chemical Processes for Cross-Domain Property Prediction Authors: Mikhail Tsitsvero, Atsuyuki Nakao, Hisaki Ikebata Published: 2025-11-26 Categories: physics.chem-ph, cs.AI, cs.DM, cs.LG arXiv: https://arxiv.org/abs/2512.05979v1 Experimental validation of chemical processes is slow and costly, limiting exploration in materials discovery. Machine learning can prioritize promising candidates, but existing data in patents and literature is heterogeneous and difficult to use. We introduce a universal directed-tree process-graph representation that unifies unstructured text, molecular structures, and numeric measurements into a single machine-readable format. To learn from this structured data, we developed a multi-modal graph neural network with a property-conditioned attention mechanism. Trained on approximately 700,000 process graphs from nearly 9,000 diverse documents, our model learns semantically rich embeddings that generalize across domains. When fine-tuned on compact, domain-specific datasets, the pretrained model achieves strong performance, demonstrating that universal process representations learned at scale transfer effectively to specialized prediction tasks with minimal additional data. --- ### Prescriptive tool for zero-emissions building fenestration design using hybrid metaheuristic algorithms Authors: Rosana Caro, Lorena Cruz, Arturo Martinez, Pablo S. Naharro, Santiago Muelas, Kevin King Sancho, Elena Cuerda, Maria del Mar Barbero-Barrera, Antonio LaTorre Published: 2025-11-26 Categories: cs.NE arXiv: https://arxiv.org/abs/2512.04102v1 Designing Zero-Emissions Buildings (ZEBs) involves balancing numerous complex objectives that traditional methods struggle to address. Fenestration, encompassing façade openings and shading systems, plays a critical role in ZEB performance due to its high thermal transmittance and solar radiation admission. This paper presents a novel simulation-based optimization method for fenestration designed for practical application. It uses a hybrid metaheuristic algorithm and relies on rules and an updatable catalog, to fully automate the design process, create a highly diverse search space, minimize biases, and generate detailed solutions ready for architectural prescription. Nineteen fenestration variables, over which architects have design flexibility, were optimized to reduce heating, cooling demand, and thermal discomfort in residential buildings. The method was tested across three Spanish climate zones. Results demonstrate that the considered optimization algorithm significantly outperforms the baseline Genetic Algorithm in both quality and robustness, with these differences proven to be statistically significant. Furthermore, the findings offer valuable insights for ZEB design, highlighting challenges in reducing cooling demand in warm climates, and showcasing the superior efficiency of automated movable shading systems compared to fixed solutions. --- ### Optimizing Life Sciences Agents in Real-Time using Reinforcement Learning Authors: Nihir Chadderwala Published: 2025-11-26 Categories: cs.LG, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2512.03065v1 Generative AI agents in life sciences face a critical challenge: determining the optimal approach for diverse queries ranging from simple factoid questions to complex mechanistic reasoning. Traditional methods rely on fixed rules or expensive labeled training data, neither of which adapts to changing conditions or user preferences. We present a novel framework that combines AWS Strands Agents with Thompson Sampling contextual bandits to enable AI agents to learn optimal decision-making strategies from user feedback alone. Our system optimizes three key dimensions: generation strategy selection (direct vs. chain-of-thought), tool selection (literature search, drug databases, etc.), and domain routing (pharmacology, molecular biology, clinical specialists). Through empirical evaluation on life science queries, we demonstrate 15-30\% improvement in user satisfaction compared to random baselines, with clear learning patterns emerging after 20-30 queries. Our approach requires no ground truth labels, adapts continuously to user preferences, and provides a principled solution to the exploration-exploitation dilemma in agentic AI systems. --- ### Globally optimized SVD compression of LLMs via Fermi-function-based rank selection and gauge fixing Authors: Roman Rausch, David Jansen, Sukhbinder Singh, Román Orús Published: 2025-11-26 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2512.03062v1 Large Language Models (LLMs) are very demanding in terms of their computational resources. Low-rank decompositions of LLM weights, e.g. via Singular Value Decomposition (SVD), is a promising approach for LLM compression, but presents several practical hurdles, e.g. selecting appropriate layer-wise ranks and getting rid of its parameter redundancy. In this work, we present two physics-inspired improvements to SVD LLM compression: (1) \textbf{FermiGrad}, a gradient-descent algorithm that determines globally optimal layer-wise ranks by relaxing the discrete singular-value truncation into a continuous optimization using the Fermi function; (2) \textbf{PivGa}, an additional \textit{lossless} compression of the low-rank factors that exploits the intrinsic gauge freedom in their parametrization. --- ### PIBNet: a Physics-Inspired Boundary Network for Multiple Scattering Simulations Authors: Rémi Marsal, Stéphanie Chaillat Published: 2025-11-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.02049v1 The boundary element method (BEM) provides an efficient numerical framework for solving multiple scattering problems in unbounded homogeneous domains, since it reduces the discretization to the domain boundaries, thereby condensing the computational complexity. The procedure first consists in determining the solution trace on the boundaries of the domain by solving a boundary integral equation, after which the volumetric solution can be recovered at low computational cost with a boundary integral representation. As the first step of the BEM represents the main computational bottleneck, we introduce PIBNet, a learning-based approach designed to approximate the solution trace. The method leverages a physics-inspired graph-based strategy to model obstacles and their long-range interactions efficiently. Then, we introduce a novel multiscale graph neural network architecture for simulating the multiple scattering. To train and evaluate our network, we present a benchmark consisting of several datasets of different types of multiple scattering problems. The results indicate that our approach not only surpasses existing state-of-the-art learning-based methods on the considered tasks but also exhibits superior generalization to settings with an increased number of obstacles. github.com/ENSTA-U2IS-AI/pibnet --- ### The Impact of Artificial Intelligence on Enterprise Decision-Making Process Authors: Ernest Górka, Dariusz Baran, Gabriela Wojak, Michał Ćwiąkała, Sebastian Zupok, Dariusz Starkowski, Dariusz Reśko, Oliwia Okrasa Published: 2025-11-26 Categories: cs.CY, cs.AI, econ.GN arXiv: https://arxiv.org/abs/2512.02048v1 Artificial intelligence improves enterprise decision-making by accelerating data analysis, reducing human error, and supporting evidence-based choices. A quantitative survey of 92 companies across multiple industries examines how AI adoption influences managerial performance, decision efficiency, and organizational barriers. Results show that 93 percent of firms use AI, primarily in customer service, data forecasting, and decision support. AI systems increase the speed and clarity of managerial decisions, yet implementation faces challenges. The most frequent barriers include employee resistance, high costs, and regulatory ambiguity. Respondents indicate that organizational factors are more significant than technological limitations. Critical competencies for successful AI use include understanding algorithmic mechanisms and change management. Technical skills such as programming play a smaller role. Employees report difficulties in adapting to AI tools, especially when formulating prompts or accepting system outputs. The study highlights the importance of integrating AI with human judgment and communication practices. When supported by adaptive leadership and transparent processes, AI adoption enhances organizational agility and strengthens decision-making performance. These findings contribute to ongoing research on how digital technologies reshape management and the evolution of hybrid human-machine decision environments. --- ### Beyond Confidence: Adaptive and Coherent Decoding for Diffusion Language Models Authors: Kecheng Chen, Ziru Liu, Xijia Tao, Hui Liu, Xinyu Fu, Suiyun Zhang, Dandan Tu, Lingpeng Kong, Rui Liu, Haoliang Li Published: 2025-11-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.02044v1 Diffusion Language Models (DLMs) have recently achieved significant success due to their any-order generation capabilities. However, existing inference methods typically rely on local, immediate-step metrics such as confidence or entropy which inherently lack a more reliable perspective. This limitation frequently leads to inconsistent sampling trajectories and suboptimal generation quality. To address this, we propose Coherent Contextual Decoding (CCD), a novel inference framework built upon two core innovations. First, CCD employs a trajectory rectification mechanism that leverages historical context to enhance sequence coherence, enabling the early rejection of suboptimal paths. We demonstrate that this mechanism is theoretically equivalent to modeling the consistency of historical steps via the conditional mutual information between context and token predictions. Building on this theoretical insight, we further address the inefficiency of conventional uniform decoding budgets. Instead of rigid allocations based on diffusion steps, we introduce an adaptive sampling strategy that dynamically adjusts the unmasking budget for each step according to our consistency metric. Consequently, our method significantly improves the quality of generation trajectories while accelerating the sampling process. Empirically, our method achieves a simultaneous enhancement in both inference speed and performance across diverse benchmarks on Dream and LLaDA, delivering up to 3.48x speedup alongside 3.91% performance improvement. --- ### HMARK: Radioactive Multi-Bit Semantic-Latent Watermarking for Diffusion Models Authors: Kexin Li, Guozhen Ding, Ilya Grishchenko, David Lie Published: 2025-11-26 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2512.00094v1 Modern generative diffusion models rely on vast training datasets, often including images with uncertain ownership or usage rights. Radioactive watermarks -- marks that transfer to a model's outputs -- can help detect when such unauthorized data has been used for training. Moreover, aside from being radioactive, an effective watermark for protecting images from unauthorized training also needs to meet other existing requirements, such as imperceptibility, robustness, and multi-bit capacity. To overcome these challenges, we propose HMARK, a novel multi-bit watermarking scheme, which encodes ownership information as secret bits in the semantic-latent space (h-space) for image diffusion models. By leveraging the interpretability and semantic significance of h-space, ensuring that watermark signals correspond to meaningful semantic attributes, the watermarks embedded by HMARK exhibit radioactivity, robustness to distortions, and minimal impact on perceptual quality. Experimental results demonstrate that HMARK achieves 98.57% watermark detection accuracy, 95.07% bit-level recovery accuracy, 100% recall rate, and 1.0 AUC on images produced by the downstream adversarial model finetuned with LoRA on watermarked data across various types of distortions. --- ### Deep Filament Extraction for 3D Concrete Printing Authors: Karam Mawas, Mehdi Maboudi, Pedro Achanccaray, Markus Gerke Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.00091v1 The architecture, engineering and construction (AEC) industry is constantly evolving to meet the demand for sustainable and effective design and construction of the built environment. In the literature, two primary deposition techniques for large-scale 3D concrete printing (3DCP) have been described, namely extrusion-based (Contour Crafting-CC) and shotcrete 3D printing (SC3DP) methods. The deposition methods use a digitally controlled nozzle to print material layer by layer. The continuous flow of concrete material used to create the printed structure is called a filament or layer. As these filaments are the essential structure defining the printed object, the filaments' geometry quality control is crucial. This paper presents an automated procedure for quality control (QC) of filaments in extrusion-based and SC3DP printing methods. The paper also describes a workflow that is independent of the sensor used for data acquisition, such as a camera, a structured light system (SLS) or a terrestrial laser scanner (TLS). This method can be used with materials in either the fresh or cured state. Thus, it can be used for online and post-printing QC. --- ### TeleViT1.0: Teleconnection-aware Vision Transformers for Subseasonal to Seasonal Wildfire Pattern Forecasts Authors: Ioannis Prapas, Nikolaos Papadopoulos, Nikolaos-Ioannis Bountos, Dimitrios Michail, Gustau Camps-Valls, Ioannis Papoutsis Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.00089v1 Forecasting wildfires weeks to months in advance is difficult, yet crucial for planning fuel treatments and allocating resources. While short-term predictions typically rely on local weather conditions, long-term forecasting requires accounting for the Earth's interconnectedness, including global patterns and teleconnections. We introduce TeleViT, a Teleconnection-aware Vision Transformer that integrates (i) fine-scale local fire drivers, (ii) coarsened global fields, and (iii) teleconnection indices. This multi-scale fusion is achieved through an asymmetric tokenization strategy that produces heterogeneous tokens processed jointly by a transformer encoder, followed by a decoder that preserves spatial structure by mapping local tokens to their corresponding prediction patches. Using the global SeasFire dataset (2001-2021, 8-day resolution), TeleViT improves AUPRC performance over U-Net++, ViT, and climatology across all lead times, including horizons up to four months. At zero lead, TeleViT with indices and global inputs reaches AUPRC 0.630 (ViT 0.617, U-Net 0.620), at 16x8day lead (around 4 months), TeleViT variants using global input maintain 0.601-0.603 (ViT 0.582, U-Net 0.578), while surpassing the climatology (0.572) at all lead times. Regional results show the highest skill in seasonally consistent fire regimes, such as African savannas, and lower skill in boreal and arid regions. Attention and attribution analyses indicate that predictions rely mainly on local tokens, with global fields and indices contributing coarse contextual information. These findings suggest that architectures explicitly encoding large-scale Earth-system context can extend wildfire predictability on subseasonal-to-seasonal timescales. --- ### Semimage: HSV-Based Semantic Image Encoding for Disentangled Text Representation Authors: Mohammad Zare Published: 2025-11-26 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.00088v2 We propose SemImage, a novel method for representing a text document as a two-dimensional semantic image to be processed by convolutional neural networks (CNNs). In a SemImage, each word is represented as a pixel in a 2D image: rows correspond to sentences and an additional boundary row is inserted between sentences to mark semantic transitions. Each pixel is not a typical RGB value but a vector in a disentangled HSV color space, encoding different linguistic features: the Hue with two components H_cos and H_sin to account for circularity encodes the topic, Saturation encodes the sentiment, and Value encodes intensity or certainty. We enforce this disentanglement via a multi-task learning framework: a ColorMapper network maps each word embedding to the HSV space, and auxiliary supervision is applied to the Hue and Saturation channels to predict topic and sentiment labels, alongside the main task objective. The insertion of dynamically computed boundary rows between sentences yields sharp visual boundaries in the image when consecutive sentences are semantically dissimilar, effectively making paragraph breaks salient. We integrate SemImage with standard 2D CNNs (e.g., ResNet) for document classification. Experiments on multi-label datasets (with both topic and sentiment annotations) and single-label benchmarks demonstrate that SemImage can achieve competitive or better accuracy than strong text classification baselines (including BERT and hierarchical attention networks) while offering enhanced interpretability. An ablation study confirms the importance of the multi-channel HSV representation and the dynamic boundary rows. Finally, we present visualizations of SemImage that qualitatively reveal clear patterns corresponding to topic shifts and sentiment changes in the generated image, suggesting that our representation makes these linguistic features visible to both humans and machines. --- ### Exploring Automated Recognition of Instructional Activity and Discourse from Multimodal Classroom Data Authors: Ivo Bueno, Ruikun Hou, Babette Bühler, Tim Fütterer, James Drimalla, Jonathan Kyle Foster, Peter Youngs, Peter Gerjets, Ulrich Trautwein, Enkelejda Kasneci Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.00087v2 Observation of classroom interactions can provide concrete feedback to teachers, but current methods rely on manual annotation, which is resource-intensive and hard to scale. This work explores AI-driven analysis of classroom recordings, focusing on multimodal instructional activity and discourse recognition as a foundation for actionable feedback. Using a densely annotated dataset of 164 hours of video and 68 lesson transcripts, we design parallel, modality-specific pipelines. For video, we evaluate zero-shot multimodal LLMs, fine-tuned vision-language models, and self-supervised video transformers on 24 activity labels. For transcripts, we fine-tune a transformer-based classifier with contextualized inputs and compare it against prompting-based LLMs on 19 discourse labels. To handle class imbalance and multi-label complexity, we apply per-label thresholding, context windows, and imbalance-aware loss functions. The results show that fine-tuned models consistently outperform prompting-based approaches, achieving macro-F1 scores of 0.577 for video and 0.460 for transcripts. These results demonstrate the feasibility of automated classroom analysis and establish a foundation for scalable teacher feedback systems. --- ### Multi-modal On-Device Learning for Monocular Depth Estimation on Ultra-low-power MCUs Authors: Davide Nadalini, Manuele Rusci, Elia Cereda, Luca Benini, Francesco Conti, Daniele Palossi Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.00086v2 Monocular depth estimation (MDE) plays a crucial role in enabling spatially-aware applications in Ultra-low-power (ULP) Internet-of-Things (IoT) platforms. However, the limited number of parameters of Deep Neural Networks for the MDE task, designed for IoT nodes, results in severe accuracy drops when the sensor data observed in the field shifts significantly from the training dataset. To address this domain shift problem, we present a multi-modal On-Device Learning (ODL) technique, deployed on an IoT device integrating a Greenwaves GAP9 MicroController Unit (MCU), a 80 mW monocular camera and a 8 x 8 pixel depth sensor, consuming $\approx$300mW. In its normal operation, this setup feeds a tiny 107 k-parameter $μ$PyD-Net model with monocular images for inference. The depth sensor, usually deactivated to minimize energy consumption, is only activated alongside the camera to collect pseudo-labels when the system is placed in a new environment. Then, the fine-tuning task is performed entirely on the MCU, using the new data. To optimize our backpropagation-based on-device training, we introduce a novel memory-driven sparse update scheme, which minimizes the fine-tuning memory to 1.2 MB, 2.2x less than a full update, while preserving accuracy (i.e., only 2% and 1.5% drops on the KITTI and NYUv2 datasets). Our in-field tests demonstrate, for the first time, that ODL for MDE can be performed in 17.8 minutes on the IoT node, reducing the root mean squared error from 4.9 to 0.6m with only 3 k self-labeled samples, collected in a real-life deployment scenario. --- ### Hyper-GoalNet: Goal-Conditioned Manipulation Policy Learning with HyperNetworks Authors: Pei Zhou, Wanting Yao, Qian Luo, Xunzhe Zhou, Yanchao Yang Published: 2025-11-26 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2512.00085v1 Goal-conditioned policy learning for robotic manipulation presents significant challenges in maintaining performance across diverse objectives and environments. We introduce Hyper-GoalNet, a framework that generates task-specific policy network parameters from goal specifications using hypernetworks. Unlike conventional methods that simply condition fixed networks on goal-state pairs, our approach separates goal interpretation from state processing -- the former determines network parameters while the latter applies these parameters to current observations. To enhance representation quality for effective policy generation, we implement two complementary constraints on the latent space: (1) a forward dynamics model that promotes state transition predictability, and (2) a distance-based constraint ensuring monotonic progression toward goal states. We evaluate our method on a comprehensive suite of manipulation tasks with varying environmental randomization. Results demonstrate significant performance improvements over state-of-the-art methods, particularly in high-variability conditions. Real-world robotic experiments further validate our method's robustness to sensor noise and physical uncertainties. Code is available at: https://github.com/wantingyao/hyper-goalnet. --- ### A Fast and Efficient Modern BERT based Text-Conditioned Diffusion Model for Medical Image Segmentation Authors: Venkata Siddharth Dhara, Pawan Kumar Published: 2025-11-26 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.00084v1 In recent times, denoising diffusion probabilistic models (DPMs) have proven effective for medical image generation and denoising, and as representation learners for downstream segmentation. However, segmentation performance is limited by the need for dense pixel-wise labels, which are expensive, time-consuming, and require expert knowledge. We propose FastTextDiff, a label-efficient diffusion-based segmentation model that integrates medical text annotations to enhance semantic representations. Our approach uses ModernBERT, a transformer capable of processing long clinical notes, to tightly link textual annotations with semantic content in medical images. Trained on MIMIC-III and MIMIC-IV, ModernBERT encodes clinical knowledge that guides cross-modal attention between visual and textual features. This study validates ModernBERT as a fast, scalable alternative to Clinical BioBERT in diffusion-based segmentation pipelines and highlights the promise of multi-modal techniques for medical image analysis. By replacing Clinical BioBERT with ModernBERT, FastTextDiff benefits from FlashAttention 2, an alternating attention mechanism, and a 2-trillion-token corpus, improving both segmentation accuracy and training efficiency over traditional diffusion-based models. --- ### Exploring Diagnostic Prompting Approach for Multimodal LLM-based Visual Complexity Assessment: A Case Study of Amazon Search Result Pages Authors: Divendar Murtadak, Yoon Kim, Trilokya Akula Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.00082v1 This study investigates whether diagnostic prompting can improve Multimodal Large Language Model (MLLM) reliability for visual complexity assessment of Amazon Search Results Pages (SRP). We compare diagnostic prompting with standard gestalt principles-based prompting using 200 Amazon SRP pages and human expert annotations. Diagnostic prompting showed notable improvements in predicting human complexity judgments, with F1-score increasing from 0.031 to 0.297 (+858\% relative improvement), though absolute performance remains modest (Cohen's $κ$ = 0.071). The decision tree revealed that models prioritize visual design elements (badge clutter: 38.6\% importance) while humans emphasize content similarity, suggesting partial alignment in reasoning patterns. Failure case analysis reveals persistent challenges in MLLM visual perception, particularly for product similarity and color intensity assessment. Our findings indicate that diagnostic prompting represents a promising initial step toward human-aligned MLLM-based evaluation, though failure cases with consistent human-MLLM disagreement require continued research and refinement in prompting approaches with larger ground truth datasets for reliable practical deployment. --- ### Digital Elevation Model Estimation from RGB Satellite Imagery using Generative Deep Learning Authors: Alif Ilham Madani, Riska A. Kuswati, Alex M. Lechner, Muhamad Risqi U. Saputra Published: 2025-11-26 Categories: eess.IV, cs.CV, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2511.21985v1 Digital Elevation Models (DEMs) are vital datasets for geospatial applications such as hydrological modeling and environmental monitoring. However, conventional methods to generate DEM, such as using LiDAR and photogrammetry, require specific types of data that are often inaccessible in resource-constrained settings. To alleviate this problem, this study proposes an approach to generate DEM from freely available RGB satellite imagery using generative deep learning, particularly based on a conditional Generative Adversarial Network (GAN). We first developed a global dataset consisting of 12K RGB-DEM pairs using Landsat satellite imagery and NASA's SRTM digital elevation data, both from the year 2000. A unique preprocessing pipeline was implemented to select high-quality, cloud-free regions and aggregate normalized RGB composites from Landsat imagery. Additionally, the model was trained in a two-stage process, where it was first trained on the complete dataset and then fine-tuned on high-quality samples filtered by Structural Similarity Index Measure (SSIM) values to improve performance on challenging terrains. The results demonstrate promising performance in mountainous regions, achieving an overall mean root-mean-square error (RMSE) of 0.4671 and a mean SSIM score of 0.2065 (scale -1 to 1), while highlighting limitations in lowland and residential areas. This study underscores the importance of meticulous preprocessing and iterative refinement in generative modeling for DEM generation, offering a cost-effective and adaptive alternative to conventional methods while emphasizing the challenge of generalization across diverse terrains worldwide. --- ### PPBoost: Progressive Prompt Boosting for Text-Driven Medical Image Segmentation Authors: Xuchen Li, Hengrui Gu, Mohan Zhang, Qin Liu, Zhen Tan, Xinyuan Zhu, Huixue Zhou, Tianlong Chen, Kaixiong Zhou Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.21984v1 Text-prompted foundation models for medical image segmentation offer an intuitive way to delineate anatomical structures from natural language queries, but their predictions often lack spatial precision and degrade under domain shift. In contrast, visual-prompted models achieve strong segmentation performance across diverse modalities by leveraging spatial cues of precise bounding-box (bbox) prompts to guide the segmentation of target lesions. However, it is costly and challenging to obtain the precise visual prompts in clinical practice. We propose PPBoost (Progressive Prompt-Boosting), a framework that bridges these limitations by transforming weak text-derived signals into strong, spatially grounded visual prompts, operating under a strict zero-shot regime with no image- or pixel-level segmentation labels. PPBoost first uses a vision-language model to produce initial pseudo-bboxes conditioned on the textual object descriptions and applies an uncertainty-aware criterion to filter unreliable predictions. The retained image-bboxes pairs are then leveraged to train a pseudo-labeled detector, producing the high-quality bboxes for the query images. During inference, PPBoost further refines the generated bboxes by appropriately expanding them to tightly cover the target anatomical structures. The enhanced spatially-grounding bbox prompts guide existing segmentation models to generate final dense masks, effectively amplifying weak text cues into strong spatial guidance. Across three datasets spanning diverse modalities and anatomies, PPBoost consistently improves Dice and Normalized Surface Distance over text- and visual-prompted baselines and, notably, surpasses few-shot segmentation models without using labeled data. PPBoost can generalize to multiple typical visual segmentation model backbones. --- ### DialBench: Towards Accurate Reading Recognition of Pointer Meter using Large Foundation Models Authors: Futian Wang, Chaoliu Weng, Xiao Wang, Zhen Chen, Zhicheng Zhao, Jin Tang Published: 2025-11-26 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.21982v1 The precise reading recognition of pointer meters plays a key role in smart power systems, but existing approaches remain fragile due to challenges like reflections, occlusions, dynamic viewing angles, and overly between thin pointers and scale markings. Up to now, this area still lacks large-scale datasets to support the development of robust algorithms. To address these challenges, this paper first presents a new large-scale benchmark dataset for dial reading, termed RPM-10K, which contains 10730 meter images that fully reflect the aforementioned key challenges. Built upon the dataset, we propose a novel vision-language model for pointer meter reading recognition, termed MRLM, based on physical relation injection. Instead of exhaustively learning image-level correlations, MRLM explicitly encodes the geometric and causal relationships between the pointer and the scale, aligning perception with physical reasoning in the spirit of world-model perspectives. Through cross-attentional fusion and adaptive expert selection, the model learns to interpret dial configurations and generate precise numeric readings. Extensive experiments fully validated the effectiveness of our proposed framework on the newly proposed benchmark dataset. Both the dataset and source code will be released on https://github.com/Event-AHU/DialBench --- ### PAT3D: Physics-Augmented Text-to-3D Scene Generation Authors: Guying Lin, Kemeng Huang, Michael Liu, Ruihan Gao, Hanke Chen, Lyuhao Chen, Beijia Lu, Taku Komura, Yuan Liu, Jun-Yan Zhu, Minchen Li Published: 2025-11-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.21978v2 We introduce PAT3D, the first physics-augmented text-to-3D scene generation framework that integrates vision-language models with physics-based simulation to produce physically plausible, simulation-ready, and intersection-free 3D scenes. Given a text prompt, PAT3D generates 3D objects, infers their spatial relations, and organizes them into a hierarchical scene tree, which is then converted into initial conditions for simulation. A differentiable rigid-body simulator ensures realistic object interactions under gravity, driving the scene toward static equilibrium without interpenetrations. To further enhance scene quality, we introduce a simulation-in-the-loop optimization procedure that guarantees physical stability and non-intersection, while improving semantic consistency with the input prompt. Experiments demonstrate that PAT3D substantially outperforms prior approaches in physical plausibility, semantic consistency, and visual quality. Beyond high-quality generation, PAT3D uniquely enables simulation-ready 3D scenes for downstream tasks such as scene editing and robotic manipulation. Code and data are available at: https://github.com/Simulation-Intelligence/PAT3D. ---