# Research Papers — 2025-12 · bundle page 61 of 175 > Full-text markdown of papers #3001–3050 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2025-12/page/61 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/60.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/62.md) ### Cartesian-nj: Extending e3nn to Irreducible Cartesian Tensor Product and Contracion Authors: Zemin Xu, Chenyu Wu, Wenbo Xie, Daiqian Xie, P. Hu Published: 2025-12-18 Categories: physics.chem-ph, cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2512.16882v1 Equivariant atomistic machine learning models have brought substantial gains in both extrapolation capability and predictive accuracy. Depending on the basis of the space, two distinct types of irreducible representations are utilized. From architectures built upon spherical tensors (STs) to more recent formulations employing irreducible Cartesian tensors (ICTs), STs have remained dominant owing to their compactness, elegance, and theoretical completeness. Nevertheless, questions have persisted regarding whether ST constructions are the only viable design principle, motivating continued development of Cartesian networks. In this work, we introduce the Cartesian-3j and Cartesian-nj symbol, which serve as direct analogues of the Wigner-3j and Wigner-nj symbol defined for tensor coupling. These coefficients enable the combination of any two ICTs into a new ICT. Building on this foundation, we extend e3nn to support irreducible Cartesian tensor product, and we release the resulting Python package as cartnn. Within this framework, we implement Cartesian counterparts of MACE, NequIP, and Allegro, allowing the first systematic comparison of Cartesian and spherical models to assess whether Cartesian formulations may offer advantages under specific conditions. Using TACE as a representative example, we further examine whether architectures constructed from irreducible Cartesian tensor product and contraction(ICTP and ICTC) are conceptually well-founded in Cartesian space and whether opportunities remain for improving their design. --- ### PolaRiS: Scalable Real-to-Sim Evaluations for Generalist Robot Policies Authors: Arhan Jain, Mingtong Zhang, Kanav Arora, William Chen, Marcel Torne, Muhammad Zubair Irshad, Sergey Zakharov, Yue Wang, Sergey Levine, Chelsea Finn, Wei-Chiu Ma, Dhruv Shah et al. Published: 2025-12-18 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2512.16881v2 A significant challenge for robot learning research is our ability to accurately measure and compare the performance of robot policies. Benchmarking in robotics is historically challenging due to the stochasticity, reproducibility, and time-consuming nature of real-world rollouts. This challenge is exacerbated for recent generalist policies, which has to be evaluated across a wide variety of scenes and tasks. Evaluation in simulation offers a scalable complement to real world evaluations, but the visual and physical domain gap between existing simulation benchmarks and the real world has made them an unreliable signal for policy improvement. Furthermore, building realistic and diverse simulated environments has traditionally required significant human effort and expertise. To bridge the gap, we introduce Policy Evaluation and Environment Reconstruction in Simulation (PolaRiS), a scalable real-to-sim framework for high-fidelity simulated robot evaluation. PolaRiS utilizes neural reconstruction methods to turn short video scans of real-world scenes into interactive simulation environments. Additionally, we develop a simple simulation data co-training recipe that bridges remaining real-to-sim gaps and enables zero-shot evaluation in unseen simulation environments. Through extensive paired evaluations between simulation and the real world, we demonstrate that PolaRiS evaluations provide a much stronger correlation to real world generalist policy performance than existing simulated benchmarks. Its simplicity also enables rapid creation of diverse simulated environments. As such, this work takes a step towards distributed and democratized evaluation for the next generation of robotic foundation models. --- ### PolaRiS: Scalable Real-to-Sim Evaluations for Generalist Robot Policies Authors: Arhan Jain, Mingtong Zhang, Kanav Arora, William Chen, Marcel Torne, Muhammad Zubair Irshad, Sergey Zakharov, Yue Wang, Sergey Levine, Chelsea Finn, Wei-Chiu Ma, Dhruv Shah et al. Published: 2025-12-18 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2512.16881v1 A significant challenge for robot learning research is our ability to accurately measure and compare the performance of robot policies. Benchmarking in robotics is historically challenging due to the stochasticity, reproducibility, and time-consuming nature of real-world rollouts. This challenge is exacerbated for recent generalist policies, which has to be evaluated across a wide variety of scenes and tasks. Evaluation in simulation offers a scalable complement to real world evaluations, but the visual and physical domain gap between existing simulation benchmarks and the real world has made them an unreliable signal for policy improvement. Furthermore, building realistic and diverse simulated environments has traditionally required significant human effort and expertise. To bridge the gap, we introduce Policy Evaluation and Environment Reconstruction in Simulation (PolaRiS), a scalable real-to-sim framework for high-fidelity simulated robot evaluation. PolaRiS utilizes neural reconstruction methods to turn short video scans of real-world scenes into interactive simulation environments. Additionally, we develop a simple simulation data co-training recipe that bridges remaining real-to-sim gaps and enables zero-shot evaluation in unseen simulation environments. Through extensive paired evaluations between simulation and the real world, we demonstrate that PolaRiS evaluations provide a much stronger correlation to real world generalist policy performance than existing simulated benchmarks. Its simplicity also enables rapid creation of diverse simulated environments. As such, this work takes a step towards distributed and democratized evaluation for the next generation of robotic foundation models. --- ### ReMeDI: Refined Memory for Disambiguation of Identities with SAM3 in Surgical Segmentation Authors: Valay Bundele, Mehran Hosseinzadeh, Hendrik P. A. Lensch Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16880v2 Accurate surgical instrument segmentation in endoscopy is crucial for computer-assisted interventions, yet remains challenging due to frequent occlusions, rapid motion, and long-term instrument re-entry. While SAM3 provides a powerful spatio-temporal framework for video object segmentation, its performance in surgical scenes is limited by indiscriminate memory updates, fixed memory capacity, and weak identity recovery after occlusions. We propose ReMeDI-SAM3, a training-free extension of SAM3, that addresses these limitations through three components: (i) relevance-aware memory filtering with a dedicated occlusion-aware memory for storing pre-occlusion frames, (ii) a piecewise interpolation scheme that expands effective memory capacity, and (iii) a feature-based re-identification module with temporal voting for reliable post-occlusion identity disambiguation. Together, these components mitigate error accumulation and enable reliable recovery after occlusions. Evaluations on EndoVis17, EndoVis18 and CholecSeg8k under a zero-shot setting show mcIoU improvements of around 5.8\%, 8\%, and 2\% respectively, over vanilla SAM3, outperforming even prior training-based approaches. --- ### Memory-Enhanced SAM3 for Occlusion-Robust Surgical Instrument Segmentation Authors: Valay Bundele, Mehran Hosseinzadeh, Hendrik P. A. Lensch Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16880v1 Accurate surgical instrument segmentation in endoscopic videos is crucial for computer-assisted interventions, yet remains challenging due to frequent occlusions, rapid motion, specular artefacts, and long-term instrument re-entry. While SAM3 provides a powerful spatio-temporal framework for video object segmentation, its performance in surgical scenes is limited by indiscriminate memory updates, fixed memory capacity, and weak identity recovery after occlusions. We propose ReMeDI-SAM3, a training-free memory-enhanced extension of SAM3, that addresses these limitations through three components: (i) relevance-aware memory filtering with a dedicated occlusion-aware memory for storing pre-occlusion frames, (ii) a piecewise interpolation scheme that expands the effective memory capacity, and (iii) a feature-based re-identification module with temporal voting for reliable post-occlusion identity disambiguation. Together, these components mitigate error accumulation and enable reliable recovery after occlusions. Evaluations on EndoVis17 and EndoVis18 under a zero-shot setting show absolute mcIoU improvements of around 7% and 16%, respectively, over vanilla SAM3, outperforming even prior training-based approaches. Project page: https://valaybundele.github.io/remedi-sam3/. --- ### Training Together, Diagnosing Better: Federated Learning for Collagen VI-Related Dystrophies Authors: Astrid Brull, Sara Aguti, Véronique Bolduc, Ying Hu, Daniel M. Jimenez-Gutierrez, Enrique Zuazua, Joaquin Del-Rio, Oleksii Sliusarenko, Haiyan Zhou, Francesco Muntoni, Carsten G. Bönnemann, Xabi Uribe-Etxebarria Published: 2025-12-18 Categories: cs.LG, cs.AI, cs.CV, cs.DC arXiv: https://arxiv.org/abs/2512.16876v1 The application of Machine Learning (ML) to the diagnosis of rare diseases, such as collagen VI-related dystrophies (COL6-RD), is fundamentally limited by the scarcity and fragmentation of available data. Attempts to expand sampling across hospitals, institutions, or countries with differing regulations face severe privacy, regulatory, and logistical obstacles that are often difficult to overcome. The Federated Learning (FL) provides a promising solution by enabling collaborative model training across decentralized datasets while keeping patient data local and private. Here, we report a novel global FL initiative using the Sherpa.ai FL platform, which leverages FL across distributed datasets in two international organizations for the diagnosis of COL6-RD, using collagen VI immunofluorescence microscopy images from patient-derived fibroblast cultures. Our solution resulted in an ML model capable of classifying collagen VI patient images into the three primary pathogenic mechanism groups associated with COL6-RD: exon skipping, glycine substitution, and pseudoexon insertion. This new approach achieved an F1-score of 0.82, outperforming single-organization models (0.57-0.75). These results demonstrate that FL substantially improves diagnostic utility and generalizability compared to isolated institutional models. Beyond enabling more accurate diagnosis, we anticipate that this approach will support the interpretation of variants of uncertain significance and guide the prioritization of sequencing strategies to identify novel pathogenic variants. --- ### Learning Confidence Ellipsoids and Applications to Robust Subspace Recovery Authors: Chao Gao, Liren Shan, Vaidehi Srinivas, Aravindan Vijayaraghavan Published: 2025-12-18 Categories: cs.DS, cs.LG, math.ST, stat.ML arXiv: https://arxiv.org/abs/2512.16875v4 We study the problem of finding confidence ellipsoids for an arbitrary distribution in high dimensions. Given samples from a distribution $D$ and a confidence parameter $α$, the goal is to find the smallest volume ellipsoid $E$ which has probability mass $\mathbb{P}_{D}(E) \ge 1-α$. Ellipsoids are a highly expressive class of confidence sets as they can capture correlations in the distribution, and can approximate any convex set. In statistics, this is the classic minimum volume estimator introduced by Rousseeuw as a robust non-parametric estimator of location and scatter. However in high dimensions, it becomes NP-hard to obtain any non-trivial approximation factor in volume when the condition number $β$ of the ellipsoid (ratio of the largest to the smallest axis length) goes to $\infty$. This motivates the focus of our paper: can we efficiently find confidence ellipsoids with volume approximation guarantees when compared to ellipsoids of bounded condition number $β$? Our main result is a polynomial time algorithm that finds an ellipsoid $E$ whose volume is within a $O(β)^{γd}$ multiplicative factor of the volume of best $β$-conditioned ellipsoid while covering at least $1-O(α/γ)$ probability mass for any $γ\in (0,1)$. In particular, setting $γ= o(1)$, this gives a $O(β)^{o(d)}$ volume approximation, with a multiplicative loss in miscoverage. We complement this with a computational hardness result that shows that such a dependence on $β$ seems necessary, even with some slack in coverage. The algorithm and analysis uses the rich primal-dual structure of the minimum volume enclosing ellipsoid and the geometric Brascamp-Lieb inequality. As a consequence, we obtain the first polynomial time algorithm with approximation guarantees on worst-case instances of the robust subspace recovery problem. --- ### Learning Confidence Ellipsoids and Applications to Robust Subspace Recovery Authors: Chao Gao, Liren Shan, Vaidehi Srinivas, Aravindan Vijayaraghavan Published: 2025-12-18 Categories: cs.DS, cs.LG, math.ST, stat.ML arXiv: https://arxiv.org/abs/2512.16875v1 We study the problem of finding confidence ellipsoids for an arbitrary distribution in high dimensions. Given samples from a distribution $D$ and a confidence parameter $α$, the goal is to find the smallest volume ellipsoid $E$ which has probability mass $\Pr_{D}(E) \ge 1-α$. Ellipsoids are a highly expressive class of confidence sets as they can capture correlations in the distribution, and can approximate any convex set. This problem has been studied in many different communities. In statistics, this is the classic minimum volume estimator introduced by Rousseeuw as a robust non-parametric estimator of location and scatter. However in high dimensions, it becomes NP-hard to obtain any non-trivial approximation factor in volume when the condition number $β$ of the ellipsoid (ratio of the largest to the smallest axis length) goes to $\infty$. This motivates the focus of our paper: can we efficiently find confidence ellipsoids with volume approximation guarantees when compared to ellipsoids of bounded condition number $β$? Our main result is a polynomial time algorithm that finds an ellipsoid $E$ whose volume is within a $O(β^{γd})$ multiplicative factor of the volume of best $β$-conditioned ellipsoid while covering at least $1-O(α/γ)$ probability mass for any $γ< α$. We complement this with a computational hardness result that shows that such a dependence seems necessary up to constants in the exponent. The algorithm and analysis uses the rich primal-dual structure of the minimum volume enclosing ellipsoid and the geometric Brascamp-Lieb inequality. As a consequence, we obtain the first polynomial time algorithm with approximation guarantees on worst-case instances of the robust subspace recovery problem. --- ### Pixel Seal: Adversarial-only training for invisible image and video watermarking Authors: Tomáš Souček, Pierre Fernandez, Hady Elsahar, Sylvestre-Alvise Rebuffi, Valeriu Lacatusu, Tuan Tran, Tom Sander, Alexandre Mourachko Published: 2025-12-18 Categories: cs.CV, cs.AI, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2512.16874v1 Invisible watermarking is essential for tracing the provenance of digital content. However, training state-of-the-art models remains notoriously difficult, with current approaches often struggling to balance robustness against true imperceptibility. This work introduces Pixel Seal, which sets a new state-of-the-art for image and video watermarking. We first identify three fundamental issues of existing methods: (i) the reliance on proxy perceptual losses such as MSE and LPIPS that fail to mimic human perception and result in visible watermark artifacts; (ii) the optimization instability caused by conflicting objectives, which necessitates exhaustive hyperparameter tuning; and (iii) reduced robustness and imperceptibility of watermarks when scaling models to high-resolution images and videos. To overcome these issues, we first propose an adversarial-only training paradigm that eliminates unreliable pixel-wise imperceptibility losses. Second, we introduce a three-stage training schedule that stabilizes convergence by decoupling robustness and imperceptibility. Third, we address the resolution gap via high-resolution adaptation, employing JND-based attenuation and training-time inference simulation to eliminate upscaling artifacts. We thoroughly evaluate the robustness and imperceptibility of Pixel Seal on different image types and across a wide range of transformations, and show clear improvements over the state-of-the-art. We finally demonstrate that the model efficiently adapts to video via temporal watermark pooling, positioning Pixel Seal as a practical and scalable solution for reliable provenance in real-world image and video settings. --- ### The Social Responsibility Stack: A Control-Theoretic Architecture for Governing Socio-Technical AI Authors: Otman A. Basir Published: 2025-12-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.16873v1 Artificial intelligence systems are increasingly deployed in domains that shape human behaviour, institutional decision-making, and societal outcomes. Existing responsible AI and governance efforts provide important normative principles but often lack enforceable engineering mechanisms that operate throughout the system lifecycle. This paper introduces the Social Responsibility Stack (SRS), a six-layer architectural framework that embeds societal values into AI systems as explicit constraints, safeguards, behavioural interfaces, auditing mechanisms, and governance processes. SRS models responsibility as a closed-loop supervisory control problem over socio-technical systems, integrating design-time safeguards with runtime monitoring and institutional oversight. We develop a unified constraint-based formulation, introduce safety-envelope and feedback interpretations, and show how fairness, autonomy, cognitive burden, and explanation quality can be continuously monitored and enforced. Case studies in clinical decision support, cooperative autonomous vehicles, and public-sector systems illustrate how SRS translates normative objectives into actionable engineering and operational controls. The framework bridges ethics, control theory, and AI governance, providing a practical foundation for accountable, adaptive, and auditable socio-technical AI systems. --- ### On the Universal Representation Property of Spiking Neural Networks Authors: Shayan Hundrieser, Philipp Tuchel, Insung Kong, Johannes Schmidt-Hieber Published: 2025-12-18 Categories: cs.NE, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2512.16872v1 Inspired by biology, spiking neural networks (SNNs) process information via discrete spikes over time, offering an energy-efficient alternative to the classical computing paradigm and classical artificial neural networks (ANNs). In this work, we analyze the representational power of SNNs by viewing them as sequence-to-sequence processors of spikes, i.e., systems that transform a stream of input spikes into a stream of output spikes. We establish the universal representation property for a natural class of spike train functions. Our results are fully quantitative, constructive, and near-optimal in the number of required weights and neurons. The analysis reveals that SNNs are particularly well-suited to represent functions with few inputs, low temporal complexity, or compositions of such functions. The latter is of particular interest, as it indicates that deep SNNs can efficiently capture composite functions via a modular design. As an application of our results, we discuss spike train classification. Overall, these results contribute to a rigorous foundation for understanding the capabilities and limitations of spike-based neuromorphic systems. --- ### Sequencing to Mitigate Catastrophic Forgetting in Continual Learning Authors: Hesham G. Moussa, Aroosa Hameed, Arashmid Akhavain Published: 2025-12-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.16871v1 To cope with real-world dynamics, an intelligent system needs to incrementally acquire, update, and exploit knowledge throughout its lifetime. This ability, known as Continual learning, provides a foundation for AI systems to develop themselves adaptively. Catastrophic forgetting is a major challenge to the progress of Continual Learning approaches, where learning a new task usually results in a dramatic performance drop on previously learned ones. Many approaches have emerged to counteract the impact of CF. Most of the proposed approaches can be categorized into five classes: replay-based, regularization-based, optimization-based, representation-based, and architecture-based. In this work, we approach the problem from a different angle, specifically by considering the optimal sequencing of tasks as they are presented to the model. We investigate the role of task sequencing in mitigating CF and propose a method for determining the optimal task order. The proposed method leverages zero-shot scoring algorithms inspired by neural architecture search (NAS). Results demonstrate that intelligent task sequencing can substantially reduce CF. Moreover, when combined with traditional continual learning strategies, sequencing offers enhanced performance and robustness against forgetting. Additionally, the presented approaches can find applications in other fields, such as curriculum learning. --- ### Semi-Supervised Online Learning on the Edge by Transforming Knowledge from Teacher Models Authors: Jiabin Xue Published: 2025-12-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.16866v1 Edge machine learning (Edge ML) enables training ML models using the vast data distributed across network edges. However, many existing approaches assume static models trained centrally and then deployed, making them ineffective against unseen data. To address this, Online Edge ML allows models to be trained directly on edge devices and updated continuously with new data. This paper explores a key challenge of Online Edge ML: "How to determine labels for truly future, unseen data points". We propose Knowledge Transformation (KT), a hybrid method combining Knowledge Distillation, Active Learning, and causal reasoning. In short, KT acts as the oracle in active learning by transforming knowledge from a teacher model to generate pseudo-labels for training a student model. To verify the validity of the method, we conducted simulation experiments with two setups: (1) using a less stable teacher model and (2) a relatively more stable teacher model. Results indicate that when a stable teacher model is given, the student model can eventually reach its expected maximum performance. KT is potentially beneficial for scenarios that meet the following circumstances: (1) when the teacher's task is generic, which means existing pre-trained models might be adequate for its task, so there will be no need to train the teacher model from scratch; and/or (2) when the label for the student's task is difficult or expensive to acquire. --- ### RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing Authors: Tianyuan Qu, Lei Ke, Xiaohang Zhan, Longxiang Tang, Yuqi Liu, Bohao Peng, Bei Yu, Dong Yu, Jiaya Jia Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16864v2 Instruction-based image editing enables natural-language control over visual modifications, yet existing models falter under Instruction-Visual Complexity (IV-Complexity), where intricate instructions meet cluttered or ambiguous scenes. We introduce RePlan (Region-aligned Planning), a plan-then-execute framework that couples a vision-language planner with a diffusion editor. The planner decomposes instructions via step-by-step reasoning and explicitly grounds them to target regions; the editor then applies changes using a training-free attention-region injection mechanism, enabling precise, parallel multi-region edits without iterative inpainting. To strengthen planning, we apply GRPO-based reinforcement learning using 1K instruction-only examples, yielding substantial gains in reasoning fidelity and format reliability. We further present IV-Edit, a benchmark focused on fine-grained grounding and knowledge-intensive edits. Across IV-Complex settings, RePlan consistently outperforms strong baselines trained on far larger datasets, improving regional precision and overall consistency. Our project page: https://replan-iv-edit.github.io --- ### RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing Authors: Tianyuan Qu, Lei Ke, Xiaohang Zhan, Longxiang Tang, Yuqi Liu, Bohao Peng, Bei Yu, Dong Yu, Jiaya Jia Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16864v1 Instruction-based image editing enables natural-language control over visual modifications, yet existing models falter under Instruction-Visual Complexity (IV-Complexity), where intricate instructions meet cluttered or ambiguous scenes. We introduce RePlan (Region-aligned Planning), a plan-then-execute framework that couples a vision-language planner with a diffusion editor. The planner decomposes instructions via step-by-step reasoning and explicitly grounds them to target regions; the editor then applies changes using a training-free attention-region injection mechanism, enabling precise, parallel multi-region edits without iterative inpainting. To strengthen planning, we apply GRPO-based reinforcement learning using 1K instruction-only examples, yielding substantial gains in reasoning fidelity and format reliability. We further present IV-Edit, a benchmark focused on fine-grained grounding and knowledge-intensive edits. Across IV-Complex settings, RePlan consistently outperforms strong baselines trained on far larger datasets, improving regional precision and overall fidelity. Our project page: https://replan-iv-edit.github.io --- ### ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning Authors: Zihan Zhou, Animesh Garg, Ajay Mandlekar, Caelan Garrett Published: 2025-12-18 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.16861v2 Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contribute to an 89% average performance increase. Finally, ReinforceGen demonstrates significant improvement through fine-tuning in our real-world evaluations. More results and videos are available at https://reinforcegen.github.io. --- ### ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning Authors: Zihan Zhou, Animesh Garg, Ajay Mandlekar, Caelan Garrett Published: 2025-12-18 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.16861v1 Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contributes to an 89% average performance increase. More results and videos available in https://reinforcegen.github.io/ --- ### Distributional AGI Safety Authors: Nenad Tomašev, Matija Franklin, Julian Jacobs, Sébastien Krier, Simon Osindero Published: 2025-12-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.16856v2 AI safety and alignment research has predominantly been focused on methods for safeguarding individual AI systems, resting on the assumption of an eventual emergence of a monolithic Artificial General Intelligence (AGI). The alternative AGI emergence hypothesis, where general capability levels are first manifested through coordination in groups of sub-AGI individual agents with complementary skills and affordances, has received far less attention. Here we argue that this patchwork AGI hypothesis needs to be given serious consideration, and should inform the development of corresponding safeguards and mitigations. The rapid deployment of advanced AI agents with tool-use capabilities and the ability to communicate and coordinate makes this an urgent safety consideration. We therefore propose a framework for distributional AGI safety that moves beyond evaluating and aligning individual agents. This framework centres on the design and implementation of virtual agentic sandbox economies (impermeable or semi-permeable), where agent-to-agent transactions are governed by robust market mechanisms, coupled with appropriate auditability, reputation management, and oversight to mitigate collective risks. --- ### Distributional AGI Safety Authors: Nenad Tomašev, Matija Franklin, Julian Jacobs, Sébastien Krier, Simon Osindero Published: 2025-12-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.16856v1 AI safety and alignment research has predominantly been focused on methods for safeguarding individual AI systems, resting on the assumption of an eventual emergence of a monolithic Artificial General Intelligence (AGI). The alternative AGI emergence hypothesis, where general capability levels are first manifested through coordination in groups of sub-AGI individual agents with complementary skills and affordances, has received far less attention. Here we argue that this patchwork AGI hypothesis needs to be given serious consideration, and should inform the development of corresponding safeguards and mitigations. The rapid deployment of advanced AI agents with tool-use capabilities and the ability to communicate and coordinate makes this an urgent safety consideration. We therefore propose a framework for distributional AGI safety that moves beyond evaluating and aligning individual agents. This framework centers on the design and implementation of virtual agentic sandbox economies (impermeable or semi-permeable), where agent-to-agent transactions are governed by robust market mechanisms, coupled with appropriate auditability, reputation management, and oversight to mitigate collective risks. --- ### TOGGLE: Temporal Logic-Guided Large Language Model Compression for Edge Authors: Khurram Khalil, Khaza Anuarul Hoque Published: 2025-12-18 Categories: cs.AI, cs.LO arXiv: https://arxiv.org/abs/2512.16855v1 Large Language Models (LLMs) deliver exceptional performance across natural language tasks but demand substantial computational resources, limiting their deployment on resource-constrained edge devices. Existing compression techniques, such as quantization and pruning, often degrade critical linguistic properties and lack formal guarantees for preserving model behavior. We propose Temporal Logic-Guided Large Language Model Compression (TOGGLE), a novel framework that leverages Signal Temporal Logic (STL) to formally specify and enforce linguistic properties during compression. TOGGLE employs an STL robustness-guided Bayesian optimization to systematically explore layer-wise quantization and pruning configurations, generating compressed models that formally satisfy specified linguistic constraints without retraining or fine-tuning. Evaluating TOGGLE on four LLM architectures (GPT-2, DeepSeek-V2 7B, LLaMA 3 8B, and Mistral 7B), we achieve up to 3.3x reduction in computational costs (FLOPs) and up to a 68.8% reduction in model size while satisfying all linguistic properties. TOGGLE represents the first integration of formal methods into LLM compression, enabling efficient, verifiable deployment of LLMs on edge hardware. --- ### GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation Authors: Amita Kamath, Kai-Wei Chang, Ranjay Krishna, Luke Zettlemoyer, Yushi Hu, Marjan Ghazvininejad Published: 2025-12-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.16853v1 Automating Text-to-Image (T2I) model evaluation is challenging; a judge model must be used to score correctness, and test prompts must be selected to be challenging for current T2I models but not the judge. We argue that satisfying these constraints can lead to benchmark drift over time, where the static benchmark judges fail to keep up with newer model capabilities. We show that benchmark drift is a significant problem for GenEval, one of the most popular T2I benchmarks. Although GenEval was well-aligned with human judgment at the time of its release, it has drifted far from human judgment over time -- resulting in an absolute error of as much as 17.7% for current models. This level of drift strongly suggests that GenEval has been saturated for some time, as we verify via a large-scale human study. To help fill this benchmarking gap, we introduce a new benchmark, GenEval 2, with improved coverage of primitive visual concepts and higher degrees of compositionality, which we show is more challenging for current models. We also introduce Soft-TIFA, an evaluation method for GenEval 2 that combines judgments for visual primitives, which we show is more well-aligned with human judgment and argue is less likely to drift from human-alignment over time (as compared to more holistic judges such as VQAScore). Although we hope GenEval 2 will provide a strong benchmark for many years, avoiding benchmark drift is far from guaranteed and our work, more generally, highlights the importance of continual audits and improvement for T2I and related automated model evaluation benchmarks. --- ### PrivateXR: Defending Privacy Attacks in Extended Reality Through Explainable AI-Guided Differential Privacy Authors: Ripan Kumar Kundu, Istiak Ahmed, Khaza Anuarul Hoque Published: 2025-12-18 Categories: cs.CR, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2512.16851v1 The convergence of artificial AI and XR technologies (AI XR) promises innovative applications across many domains. However, the sensitive nature of data (e.g., eye-tracking) used in these systems raises significant privacy concerns, as adversaries can exploit these data and models to infer and leak personal information through membership inference attacks (MIA) and re-identification (RDA) with a high success rate. Researchers have proposed various techniques to mitigate such privacy attacks, including differential privacy (DP). However, AI XR datasets often contain numerous features, and applying DP uniformly can introduce unnecessary noise to less relevant features, degrade model accuracy, and increase inference time, limiting real-time XR deployment. Motivated by this, we propose a novel framework combining explainable AI (XAI) and DP-enabled privacy-preserving mechanisms to defend against privacy attacks. Specifically, we leverage post-hoc explanations to identify the most influential features in AI XR models and selectively apply DP to those features during inference. We evaluate our XAI-guided DP approach on three state-of-the-art AI XR models and three datasets: cybersickness, emotion, and activity classification. Our results show that the proposed method reduces MIA and RDA success rates by up to 43% and 39%, respectively, for cybersickness tasks while preserving model utility with up to 97% accuracy using Transformer models. Furthermore, it improves inference time by up to ~2x compared to traditional DP approaches. To demonstrate practicality, we deploy the XAI-guided DP AI XR models on an HTC VIVE Pro headset and develop a user interface (UI), namely PrivateXR, allowing users to adjust privacy levels (e.g., low, medium, high) while receiving real-time task predictions, protecting user privacy during XR gameplay. --- ### Meta-RL Induces Exploration in Language Agents Authors: Yulun Jiang, Liangze Jiang, Damien Teney, Michael Moor, Maria Brbic Published: 2025-12-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.16848v2 Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agents to actively explore and learn from the environment feedback at test time. LaMer consists of two key components: (i) a cross-episode training framework to encourage exploration and long-term rewards optimization; and (ii) in-context policy adaptation via reflection, allowing the agent to adapt their policy from task feedback signal without gradient update. Experiments across diverse environments show that LaMer significantly improves performance over RL baselines, with 11%, 14%, and 19% performance gains on Sokoban, MineSweeper and Webshop, respectively. Moreover, LaMer also demonstrates better generalization to more challenging or previously unseen tasks compared to the RL-trained agents. Overall, our results demonstrate that Meta-RL provides a principled approach to induce exploration in language agents, enabling more robust adaptation to novel environments through learned exploration strategies. --- ### Meta-RL Induces Exploration in Language Agents Authors: Yulun Jiang, Liangze Jiang, Damien Teney, Michael Moor, Maria Brbic Published: 2025-12-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.16848v1 Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agents to actively explore and learn from the environment feedback at test time. LaMer consists of two key components: (i) a cross-episode training framework to encourage exploration and long-term rewards optimization; and (ii) in-context policy adaptation via reflection, allowing the agent to adapt their policy from task feedback signal without gradient update. Experiments across diverse environments show that LaMer significantly improves performance over RL baselines, with 11%, 14%, and 19% performance gains on Sokoban, MineSweeper and Webshop, respectively. Moreover, LaMer also demonstrates better generalization to more challenging or previously unseen tasks compared to the RL-trained agents. Overall, our results demonstrate that Meta-RL provides a principled approach to induce exploration in language agents, enabling more robust adaptation to novel environments through learned exploration strategies. --- ### LLMCache: Layer-Wise Caching Strategies for Accelerated Reuse in Transformer Inference Authors: Harsh Vardhan Bansal Published: 2025-12-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.16843v1 Transformer-based language models have achieved remarkable performance across a wide range of tasks, yet their high inference latency poses a significant challenge for real-timeand large-scale deployment. While existing caching mechanisms,such as token-level key-value caches, offer speedups in autore-gressive decoding, they are limited in scope and applicability. In this paper, we present LLMCache, a novel layer-wise caching framework that accelerates transformer inference by reusing intermediate activations based on semantic similarity of input sequences. Unlike prior work, LLMCache is model-agnostic,operates across both encoder and decoder architectures, and supports caching at arbitrary transformer layers. We introduce a lightweight fingerprinting mechanism for matching seman-tically similar inputs and propose adaptive eviction strategies to manage cache staleness. Experiments on BERT and GPT-2 across SQuAD, WikiText-103, and OpenBookQA show up to 3.1 X speedup in inference time with <0.5% accuracy degradation. Our results highlight LLMCache as a practical and general-purpose solution for optimizing transformer inference in real-world applications --- ### OPENTOUCH: Bringing Full-Hand Touch to Real-World Interaction Authors: Yuxin Ray Song, Jinzhou Li, Rao Fu, Devin Murphy, Kaichen Zhou, Rishi Shiv, Yaqi Li, Haoyu Xiong, Crystal Elaine Owens, Yilun Du, Yiyue Luo, Xianyi Cheng et al. Published: 2025-12-18 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2512.16842v1 The human hand is our primary interface to the physical world, yet egocentric perception rarely knows when, where, or how forcefully it makes contact. Robust wearable tactile sensors are scarce, and no existing in-the-wild datasets align first-person video with full-hand touch. To bridge the gap between visual perception and physical interaction, we present OpenTouch, the first in-the-wild egocentric full-hand tactile dataset, containing 5.1 hours of synchronized video-touch-pose data and 2,900 curated clips with detailed text annotations. Using OpenTouch, we introduce retrieval and classification benchmarks that probe how touch grounds perception and action. We show that tactile signals provide a compact yet powerful cue for grasp understanding, strengthen cross-modal alignment, and can be reliably retrieved from in-the-wild video queries. By releasing this annotated vision-touch-pose dataset and benchmark, we aim to advance multimodal egocentric perception, embodied learning, and contact-rich robotic manipulation. --- ### Radiology Report Generation with Layer-Wise Anatomical Attention Authors: Emmanuel D. Muñiz-De-León, Jorge A. Rosales-de-Golferichs, Ana S. Muñoz-Rodríguez, Alejandro I. Trejo-Castro, Eduardo de Avila-Armenta, Antonio Martínez-Torteya Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16841v1 Automatic radiology report generation is a promising application of multimodal deep learning, aiming to reduce reporting workload and improve consistency. However, current state-of-the-art (SOTA) systems - such as Multimodal AI for Radiology Applications (MAIRA-2) and Medical Pathways Language Model-Multimodal (MedPaLM-M) - depend on large-scale multimodal training, clinical metadata, and multiple imaging views, making them resource-intensive and inaccessible for most settings. We introduce a compact image-to-text architecture that generates the Findings section of chest X-ray reports from a single frontal image. The model combines a frozen Self-Distillation with No Labels v3 (DINOv3) Vision Transformer (ViT) encoder with a Generative Pre-trained Transformer 2 (GPT-2) decoder enhanced by layer-wise anatomical attention. This mechanism integrates lung and heart segmentation masks through hierarchical Gaussian smoothing, biasing attention toward clinically relevant regions without adding trainable parameters. Evaluated on the official Medical Information Mart for Intensive Care-Chest X-ray (MIMIC-CXR) dataset using Chest Radiograph Expert (CheXpert) and Radiology Graph (RadGraph) metrics, our approach achieved substantial gains: CheXpert Macro-F1 for five key pathologies increased by 168% (0.083 -> 0.238) and Micro-F1 by 146% (0.137 -> 0.337), while broader performance across 14 observations improved by 86% (0.170 -> 0.316). Structural coherence also improved, with RadGraph F1 rising by 9.7%. Despite its small size and purely image-conditioned design, the model demonstrates that decoder-level anatomical guidance improves spatial grounding and enhances coherence in clinically relevant regions. The source code is publicly available at: https://github.com/devMuniz02/UDEM-CXR-Reporting-Thesis-2025. --- ### What Do Prosody and Text Convey? Characterizing How Meaningful Information is Distributed Across Multiple Channels Authors: Aditya Yadavalli, Tiago Pimentel, Tamar I Regev, Ethan Wilcox, Alex Warstadt Published: 2025-12-18 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.16832v1 Prosody -- the melody of speech -- conveys critical information often not captured by the words or text of a message. In this paper, we propose an information-theoretic approach to quantify how much information is expressed by prosody alone and not by text, and crucially, what that information is about. Our approach applies large speech and language models to estimate the mutual information between a particular dimension of an utterance's meaning (e.g., its emotion) and any of its communication channels (e.g., audio or text). We then use this approach to quantify how much information is conveyed by audio and text about sarcasm, emotion, and questionhood, using speech from television and podcasts. We find that for sarcasm and emotion the audio channel -- and by implication the prosodic channel -- transmits over an order of magnitude more information about these features than the text channel alone, at least when long-term context beyond the current sentence is unavailable. For questionhood, prosody provides comparatively less additional information. We conclude by outlining a program applying our approach to more dimensions of meaning, communication channels, and languages. --- ### Next-Generation License Plate Detection and Recognition System using YOLOv8 Authors: Arslan Amin, Rafia Mumtaz, Muhammad Jawad Bashir, Syed Mohammad Hassan Zaidi Published: 2025-12-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.16826v1 In the evolving landscape of traffic management and vehicle surveillance, efficient license plate detection and recognition are indispensable. Historically, many methodologies have tackled this challenge, but consistent real-time accuracy, especially in diverse environments, remains elusive. This study examines the performance of YOLOv8 variants on License Plate Recognition (LPR) and Character Recognition tasks, crucial for advancing Intelligent Transportation Systems. Two distinct datasets were employed for training and evaluation, yielding notable findings. The YOLOv8 Nano variant demonstrated a precision of 0.964 and mAP50 of 0.918 on the LPR task, while the YOLOv8 Small variant exhibited a precision of 0.92 and mAP50 of 0.91 on the Character Recognition task. A custom method for character sequencing was introduced, effectively sequencing the detected characters based on their x-axis positions. An optimized pipeline, utilizing YOLOv8 Nano for LPR and YOLOv8 Small for Character Recognition, is proposed. This configuration not only maintains computational efficiency but also ensures high accuracy, establishing a robust foundation for future real-world deployments on edge devices within Intelligent Transportation Systems. This effort marks a significant stride towards the development of smarter and more efficient urban infrastructures. --- ### Tiny Recursive Control: Iterative Reasoning for Efficient Optimal Control Authors: Amit Jain, Richard Linares Published: 2025-12-18 Categories: cs.LG, math.DS arXiv: https://arxiv.org/abs/2512.16824v1 Neural network controllers increasingly demand millions of parameters, and language model approaches push into the billions. For embedded aerospace systems with strict power and latency constraints, this scaling is prohibitive. We present Tiny Recursive Control (TRC), a neural architecture based on a counterintuitive principle: capacity can emerge from iteration depth rather than parameter count. TRC applies compact networks (approximately 1.5M parameters) repeatedly through a two-level hierarchical latent structure, refining control sequences by simulating trajectories and correcting based on tracking error. Because the same weights process every refinement step, adding iterations increases computation without increasing memory. We evaluate TRC on nonlinear control problems including oscillator stabilization and powered descent with fuel constraints. Across these domains, TRC achieves near-optimal control costs while requiring only millisecond-scale inference on GPU and under 10~MB memory, two orders of magnitude smaller than language model baselines. These results demonstrate that recursive reasoning, previously confined to discrete tasks, transfers effectively to continuous control synthesis. --- ### MEPIC: Memory Efficient Position Independent Caching for LLM Serving Authors: Qian Wang, Zahra Yousefijamarani, Morgan Lindsay Heisler, Rongzhi Gu, Bai Xiaolong, Shan Yizhou, Wei Zhang, Wang Lan, Ying Xiong, Yong Zhang, Zhenan Fan Published: 2025-12-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.16822v1 Modern LLM applications such as deep-research assistants, coding agents, and Retrieval-Augmented Generation (RAG) systems, repeatedly process long prompt histories containing shared document or code chunks, creating significant pressure on the Key Value (KV) cache, which must operate within limited memory while sustaining high throughput and low latency. Prefix caching partially alleviates some of these costs by reusing KV cache for previously processed tokens, but limited by strict prefix matching. Position-independent caching (PIC) enables chunk-level reuse at arbitrary positions, but requires selective recomputation and positional-encoding (PE) adjustments. However, because these operations vary across queries, KV for the same chunk diverges across requests. Moreover, without page alignment, chunk KV layouts diverge in memory, preventing page sharing. These issues result in only modest HBM savings even when many requests reuse the same content. We present MEPIC, a memory-efficient PIC system that enables chunk KV reuse across positions, requests, and batches. MEPIC aligns chunk KV to paged storage, shifts recomputation from token- to block-level so only the first block is request-specific, removes positional encodings via Rotary Position Embedding (RoPE) fusion in the attention kernel, and makes remaining blocks fully shareable. These techniques eliminate most duplicate chunk KV in HBM, reducing usage by up to 2x over state-of-the-art PIC at comparable latency and accuracy, and up to 5x for long prompts, without any model changes. --- ### DenseBEV: Transforming BEV Grid Cells into 3D Objects Authors: Marius Dähling, Sebastian Krebs, J. Marius Zöllner Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16818v1 In current research, Bird's-Eye-View (BEV)-based transformers are increasingly utilized for multi-camera 3D object detection. Traditional models often employ random queries as anchors, optimizing them successively. Recent advancements complement or replace these random queries with detections from auxiliary networks. We propose a more intuitive and efficient approach by using BEV feature cells directly as anchors. This end-to-end approach leverages the dense grid of BEV queries, considering each cell as a potential object for the final detection task. As a result, we introduce a novel two-stage anchor generation method specifically designed for multi-camera 3D object detection. To address the scaling issues of attention with a large number of queries, we apply BEV-based Non-Maximum Suppression, allowing gradients to flow only through non-suppressed objects. This ensures efficient training without the need for post-processing. By using BEV features from encoders such as BEVFormer directly as object queries, temporal BEV information is inherently embedded. Building on the temporal BEV information already embedded in our object queries, we introduce a hybrid temporal modeling approach by integrating prior detections to further enhance detection performance. Evaluating our method on the nuScenes dataset shows consistent and significant improvements in NDS and mAP over the baseline, even with sparser BEV grids and therefore fewer initial anchors. It is particularly effective for small objects, enhancing pedestrian detection with a 3.8% mAP increase on nuScenes and an 8% increase in LET-mAP on Waymo. Applying our method, named DenseBEV, to the challenging Waymo Open dataset yields state-of-the-art performance, achieving a LET-mAP of 60.7%, surpassing the previous best by 5.4%. Code is available at https://github.com/mdaehl/DenseBEV. --- ### Grammar-Forced Translation of Natural Language to Temporal Logic using LLMs Authors: William English, Dominic Simon, Sumit Kumar Jha, Rickard Ewetz Published: 2025-12-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.16814v1 Translating natural language (NL) into a formal language such as temporal logic (TL) is integral for human communication with robots and autonomous systems. State-of-the-art approaches decompose the task into a lifting of atomic propositions (APs) phase and a translation phase. However, existing methods struggle with accurate lifting, the existence of co-references, and learning from limited data. In this paper, we propose a framework for NL to TL translation called Grammar Forced Translation (GraFT). The framework is based on the observation that previous work solves both the lifting and translation steps by letting a language model iteratively predict tokens from its full vocabulary. In contrast, GraFT reduces the complexity of both tasks by restricting the set of valid output tokens from the full vocabulary to only a handful in each step. The solution space reduction is obtained by exploiting the unique properties of each problem. We also provide a theoretical justification for why the solution space reduction leads to more efficient learning. We evaluate the effectiveness of GraFT using the CW, GLTL, and Navi benchmarks. Compared with state-of-the-art translation approaches, it can be observed that GraFT the end-to-end translation accuracy by 5.49% and out-of-domain translation accuracy by 14.06% on average. --- ### Coordinated Anti-Jamming Resilience in Swarm Networks via Multi-Agent Reinforcement Learning Authors: Bahman Abolhassani, Tugba Erpek, Kemal Davaslioglu, Yalin E. Sagduyu, Sastry Kompella Published: 2025-12-18 Categories: cs.NI, cs.AI, cs.DC, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2512.16813v1 Reactive jammers pose a severe security threat to robotic-swarm networks by selectively disrupting inter-agent communications and undermining formation integrity and mission success. Conventional countermeasures such as fixed power control or static channel hopping are largely ineffective against such adaptive adversaries. This paper presents a multi-agent reinforcement learning (MARL) framework based on the QMIX algorithm to improve the resilience of swarm communications under reactive jamming. We consider a network of multiple transmitter-receiver pairs sharing channels while a reactive jammer with Markovian threshold dynamics senses aggregate power and reacts accordingly. Each agent jointly selects transmit frequency (channel) and power, and QMIX learns a centralized but factorizable action-value function that enables coordinated yet decentralized execution. We benchmark QMIX against a genie-aided optimal policy in a no-channel-reuse setting, and against local Upper Confidence Bound (UCB) and a stateless reactive policy in a more general fading regime with channel reuse enabled. Simulation results show that QMIX rapidly converges to cooperative policies that nearly match the genie-aided bound, while achieving higher throughput and lower jamming incidence than the baselines, thereby demonstrating MARL's effectiveness for securing autonomous swarms in contested environments. --- ### GeoPredict: Leveraging Predictive Kinematics and 3D Gaussian Geometry for Precise VLA Manipulation Authors: Jingjing Qian, Boyao Han, Chen Shi, Lei Xiao, Long Yang, Shaoshuai Shi, Li Jiang Published: 2025-12-18 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2512.16811v2 Vision-Language-Action (VLA) models achieve strong generalization in robotic manipulation but remain largely reactive and 2D-centric, making them unreliable in tasks that require precise 3D reasoning. We propose GeoPredict, a geometry-aware VLA framework that augments a continuous-action policy with predictive kinematic and geometric priors. GeoPredict introduces a trajectory-level module that encodes motion history and predicts multi-step 3D keypoint trajectories of robot arms, and a predictive 3D Gaussian geometry module that forecasts workspace geometry with track-guided refinement along future keypoint trajectories. These predictive modules serve exclusively as training-time supervision through depth-based rendering, while inference requires only lightweight additional query tokens without invoking any 3D decoding. Experiments on RoboCasa Human-50, LIBERO, and real-world manipulation tasks show that GeoPredict consistently outperforms strong VLA baselines, especially in geometry-intensive and spatially demanding scenarios. --- ### GeoPredict: Leveraging Predictive Kinematics and 3D Gaussian Geometry for Precise VLA Manipulation Authors: Jingjing Qian, Boyao Han, Chen Shi, Lei Xiao, Long Yang, Shaoshuai Shi, Li Jiang Published: 2025-12-18 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2512.16811v1 Vision-Language-Action (VLA) models achieve strong generalization in robotic manipulation but remain largely reactive and 2D-centric, making them unreliable in tasks that require precise 3D reasoning. We propose GeoPredict, a geometry-aware VLA framework that augments a continuous-action policy with predictive kinematic and geometric priors. GeoPredict introduces a trajectory-level module that encodes motion history and predicts multi-step 3D keypoint trajectories of robot arms, and a predictive 3D Gaussian geometry module that forecasts workspace geometry with track-guided refinement along future keypoint trajectories. These predictive modules serve exclusively as training-time supervision through depth-based rendering, while inference requires only lightweight additional query tokens without invoking any 3D decoding. Experiments on RoboCasa Human-50, LIBERO, and real-world manipulation tasks show that GeoPredict consistently outperforms strong VLA baselines, especially in geometry-intensive and spatially demanding scenarios. --- ### Exploration of Augmentation Strategies in Multi-modal Retrieval-Augmented Generation for the Biomedical Domain: A Case Study Evaluating Question Answering in Glycobiology Authors: Primož Kocbek, Azra Frkatović-Hodžić, Dora Lalić, Vivian Hui, Gordan Lauc, Gregor Štiglic Published: 2025-12-18 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.16802v1 Multi-modal retrieval-augmented generation (MM-RAG) promises grounded biomedical QA, but it is unclear when to (i) convert figures/tables into text versus (ii) use optical character recognition (OCR)-free visual retrieval that returns page images and leaves interpretation to the generator. We study this trade-off in glycobiology, a visually dense domain. We built a benchmark of 120 multiple-choice questions (MCQs) from 25 papers, stratified by retrieval difficulty (easy text, medium figures/tables, hard cross-evidence). We implemented four augmentations-None, Text RAG, Multi-modal conversion, and late-interaction visual retrieval (ColPali)-using Docling parsing and Qdrant indexing. We evaluated mid-size open-source and frontier proprietary models (e.g., Gemma-3-27B-IT, GPT-4o family). Additional testing used the GPT-5 family and multiple visual retrievers (ColPali/ColQwen/ColFlor). Accuracy with Agresti-Coull 95% confidence intervals (CIs) was computed over 5 runs per configuration. With Gemma-3-27B-IT, Text and Multi-modal augmentation outperformed OCR-free retrieval (0.722-0.740 vs. 0.510 average accuracy). With GPT-4o, Multi-modal achieved 0.808, with Text 0.782 and ColPali 0.745 close behind; within-model differences were small. In follow-on experiments with the GPT-5 family, the best results with ColPali and ColFlor improved by ~2% to 0.828 in both cases. In general, across the GPT-5 family, ColPali, ColQwen, and ColFlor were statistically indistinguishable. GPT-5-nano trailed larger GPT-5 variants by roughly 8-10%. Pipeline choice is capacity-dependent: converting visuals to text lowers the reader burden and is more reliable for mid-size models, whereas OCR-free visual retrieval becomes competitive under frontier models. Among retrievers, ColFlor offers parity with heavier options at a smaller footprint, making it an efficient default when strong generators are available. --- ### From Facts to Conclusions : Integrating Deductive Reasoning in Retrieval-Augmented LLMs Authors: Shubham Mishra, Samyek Jain, Gorang Mehrishi, Shiv Tiwari, Harsh Sharma, Pratik Narang, Dhruv Kumar Published: 2025-12-18 Categories: cs.CL, cs.AI, cs.CY, cs.IR arXiv: https://arxiv.org/abs/2512.16795v1 Retrieval-Augmented Generation (RAG) grounds large language models (LLMs) in external evidence, but fails when retrieved sources conflict or contain outdated or subjective information. Prior work address these issues independently but lack unified reasoning supervision. We propose a reasoning-trace-augmented RAG framework that adds structured, interpretable reasoning across three stages : (1) document-level adjudication, (2) conflict analysis, and (3) grounded synthesis, producing citation-linked answers or justified refusals. A Conflict-Aware Trust-Score (CATS) pipeline is introduced which evaluates groundedness, factual correctness, refusal accuracy, and conflict-behavior alignment using an LLM-as-a-Judge. Our 539-query reasoning dataset and evaluation pipeline establish a foundation for conflict-aware, interpretable RAG systems. Experimental results demonstrate substantial gains over baselines, most notably with Qwen, where Supervised Fine-Tuning improved End-to-End answer correctness from 0.069 to 0.883 and behavioral adherence from 0.074 to 0.722. --- ### Delay-Aware Multi-Stage Edge Server Upgrade with Budget Constraint Authors: Endar Suprih Wihidayat, Sieteng Soh, Kwan-Wu Chin, Duc-son Pham Published: 2025-12-18 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2512.16792v1 In this paper, the Multi-stage Edge Server Upgrade (M-ESU) is proposed as a new network planning problem, involving the upgrading of an existing multi-access edge computing (MEC) system through multiple stages (e.g., over several years). More precisely, the problem considers two key decisions: (i) whether to deploy additional edge servers or upgrade those already installed, and (ii) how tasks should be offloaded so that the average number of tasks that meet their delay requirement is maximized. The framework specifically involves: (i) deployment of new servers combined with capacity upgrades for existing servers, and (ii) the optimal task offloading to maximize the average number of tasks with a delay requirement. It also considers the following constraints: (i) budget per stage, (ii) server deployment and upgrade cost (in $) and cost depreciation rate, (iii) computation resource of servers, (iv) number of tasks and their growth rate (in %), and (v) the increase in task sizes and stricter delay requirements over time. We present two solutions: a Mixed Integer Linear Programming (MILP) model and an efficient heuristic algorithm (M-ESU/H). MILP yields the optimal solution for small networks, whereas M-ESU/H is used in large-scale networks. For small networks, the simulation results show that the solution computed by M-ESU/H is within 1.25% of the optimal solution while running several orders of magnitude faster. For large networks, M-ESU/H is compared against three alternative heuristic solutions that consider only server deployment, or giving priority to server deployment or upgrade. Our experiments show that M-ESU/H yields up to 21.57% improvement in task satisfaction under identical budget and demand growth conditions, confirming its scalability and practical value for long-term MEC systems. --- ### KineST: A Kinematics-guided Spatiotemporal State Space Model for Human Motion Tracking from Sparse Signals Authors: Shuting Zhao, Zeyu Xiao, Xinrong Chen Published: 2025-12-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.16791v1 Full-body motion tracking plays an essential role in AR/VR applications, bridging physical and virtual interactions. However, it is challenging to reconstruct realistic and diverse full-body poses based on sparse signals obtained by head-mounted displays, which are the main devices in AR/VR scenarios. Existing methods for pose reconstruction often incur high computational costs or rely on separately modeling spatial and temporal dependencies, making it difficult to balance accuracy, temporal coherence, and efficiency. To address this problem, we propose KineST, a novel kinematics-guided state space model, which effectively extracts spatiotemporal dependencies while integrating local and global pose perception. The innovation comes from two core ideas. Firstly, in order to better capture intricate joint relationships, the scanning strategy within the State Space Duality framework is reformulated into kinematics-guided bidirectional scanning, which embeds kinematic priors. Secondly, a mixed spatiotemporal representation learning approach is employed to tightly couple spatial and temporal contexts, balancing accuracy and smoothness. Additionally, a geometric angular velocity loss is introduced to impose physically meaningful constraints on rotational variations for further improving motion stability. Extensive experiments demonstrate that KineST has superior performance in both accuracy and temporal consistency within a lightweight framework. Project page: https://kaka-1314.github.io/KineST/ --- ### Few-Shot Specific Emitter Identification via Integrated Complex Variational Mode Decomposition and Spatial Attention Transfer Authors: Chenyu Zhu, Zeyang Li, Ziyi Xie, Jie Zhang Published: 2025-12-18 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2512.16786v2 Specific emitter identification (SEI) utilizes passive hardware characteristics to authenticate transmitters, providing a robust physical-layer security solution. However, most deep-learning-based methods rely on extensive data or require prior information, which poses challenges in real-world scenarios with limited labeled data. We propose an integrated complex variational mode decomposition algorithm that decomposes and reconstructs complex-valued signals to approximate the original transmitted signals, thereby enabling more accurate feature extraction. We further utilize a temporal convolutional network to effectively model the sequential signal characteristics, and introduce a spatial attention mechanism to adaptively weight informative signal segments, significantly enhancing identification performance. Additionally, the branch network allows leveraging pre-trained weights from other data while reducing the need for auxiliary datasets. Ablation experiments on the simulated data demonstrate the effectiveness of each component of the model. An accuracy comparison on a public dataset reveals that our method achieves 96% accuracy using only 10 symbols without requiring any prior knowledge. --- ### Few-Shot Specific Emitter Identification via Integrated Complex Variational Mode Decomposition and Spatial Attention Transfer Authors: Chenyu Zhu, Zeyang Li, Ziyi Xie, Jie Zhang Published: 2025-12-18 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2512.16786v1 Specific emitter identification (SEI) utilizes passive hardware characteristics to authenticate transmitters, providing a robust physical-layer security solution. However, most deep-learning-based methods rely on extensive data or require prior information, which poses challenges in real-world scenarios with limited labeled data. We propose an integrated complex variational mode decomposition algorithm that decomposes and reconstructs complex-valued signals to approximate the original transmitted signals, thereby enabling more accurate feature extraction. We further utilize a temporal convolutional network to effectively model the sequential signal characteristics, and introduce a spatial attention mechanism to adaptively weight informative signal segments, significantly enhancing identification performance. Additionally, the branch network allows leveraging pre-trained weights from other data while reducing the need for auxiliary datasets. Ablation experiments on the simulated data demonstrate the effectiveness of each component of the model. An accuracy comparison on a public dataset reveals that our method achieves 96% accuracy using only 10 symbols without requiring any prior knowledge. --- ### R3ST: A Synthetic 3D Dataset With Realistic Trajectories Authors: Simone Teglia, Claudia Melis Tonti, Francesco Pro, Leonardo Russo, Andrea Alfarano, Leonardo Pentassuglia, Irene Amerini Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16784v1 Datasets are essential to train and evaluate computer vision models used for traffic analysis and to enhance road safety. Existing real datasets fit real-world scenarios, capturing authentic road object behaviors, however, they typically lack precise ground-truth annotations. In contrast, synthetic datasets play a crucial role, allowing for the annotation of a large number of frames without additional costs or extra time. However, a general drawback of synthetic datasets is the lack of realistic vehicle motion, since trajectories are generated using AI models or rule-based systems. In this work, we introduce R3ST (Realistic 3D Synthetic Trajectories), a synthetic dataset that overcomes this limitation by generating a synthetic 3D environment and integrating real-world trajectories derived from SinD, a bird's-eye-view dataset recorded from drone footage. The proposed dataset closes the gap between synthetic data and realistic trajectories, advancing the research in trajectory forecasting of road vehicles, offering both accurate multimodal ground-truth annotations and authentic human-driven vehicle trajectories. --- ### Towards Mass Spectrum Analysis with ASP Authors: Nils Küchenmeister, Alex Ivliev, Markus Krötzsch Published: 2025-12-18 Categories: cs.LO, cs.AI arXiv: https://arxiv.org/abs/2512.16780v1 We present a new use of Answer Set Programming (ASP) to discover the molecular structure of chemical samples based on the relative abundance of elements and structural fragments, as measured in mass spectrometry. To constrain the exponential search space for this combinatorial problem, we develop canonical representations of molecular structures and an ASP implementation that uses these definitions. We evaluate the correctness of our implementation over a large set of known molecular structures, and we compare its quality and performance to other ASP symmetry-breaking methods and to a commercial tool from analytical chemistry. Under consideration in Theory and Practice of Logic Programming (TPLP). --- ### Non-Linear Strong Data-Processing for Quantum Hockey-Stick Divergences Authors: Theshani Nuradha, Ian George, Christoph Hirche Published: 2025-12-18 Categories: quant-ph, cs.CR, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2512.16778v1 Data-processing is a desired property of classical and quantum divergences and information measures. In information theory, the contraction coefficient measures how much the distinguishability of quantum states decreases when they are transmitted through a quantum channel, establishing linear strong data-processing inequalities (SDPI). However, these linear SDPI are not always tight and can be improved in most of the cases. In this work, we establish non-linear SDPI for quantum hockey-stick divergence for noisy channels that satisfy a certain noise criterion. We also note that our results improve upon existing linear SDPI for quantum hockey-stick divergences and also non-linear SDPI for classical hockey-stick divergence. We define $F_γ$ curves generalizing Dobrushin curves for the quantum setting while characterizing SDPI for the sequential composition of heterogeneous channels. In addition, we derive reverse-Pinsker type inequalities for $f$-divergences with additional constraints on hockey-stick divergences. We show that these non-linear SDPI can establish tighter finite mixing times that cannot be achieved through linear SDPI. Furthermore, we find applications of these in establishing stronger privacy guarantees for the composition of sequential private quantum channels when privacy is quantified by quantum local differential privacy. --- ### Kling-Omni Technical Report Authors: Kling Team, Jialu Chen, Yuanzheng Ci, Xiangyu Du, Zipeng Feng, Kun Gai, Sainan Guo, Feng Han, Jingbin He, Kang He, Xiao Hu, Xiaohua Hu et al. Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16776v1 We present Kling-Omni, a generalist generative framework designed to synthesize high-fidelity videos directly from multimodal visual language inputs. Adopting an end-to-end perspective, Kling-Omni bridges the functional separation among diverse video generation, editing, and intelligent reasoning tasks, integrating them into a holistic system. Unlike disjointed pipeline approaches, Kling-Omni supports a diverse range of user inputs, including text instructions, reference images, and video contexts, processing them into a unified multimodal representation to deliver cinematic-quality and highly-intelligent video content creation. To support these capabilities, we constructed a comprehensive data system that serves as the foundation for multimodal video creation. The framework is further empowered by efficient large-scale pre-training strategies and infrastructure optimizations for inference. Comprehensive evaluations reveal that Kling-Omni demonstrates exceptional capabilities in in-context generation, reasoning-based editing, and multimodal instruction following. Moving beyond a content creation tool, we believe Kling-Omni is a pivotal advancement toward multimodal world simulators capable of perceiving, reasoning, generating and interacting with the dynamic and complex worlds. --- ### FlowDet: Unifying Object Detection and Generative Transport Flows Authors: Enis Baty, C. P. Bridges, Simon Hadfield Published: 2025-12-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.16771v1 We present FlowDet, the first formulation of object detection using modern Conditional Flow Matching techniques. This work follows from DiffusionDet, which originally framed detection as a generative denoising problem in the bounding box space via diffusion. We revisit and generalise this formulation to a broader class of generative transport problems, while maintaining the ability to vary the number of boxes and inference steps without re-training. In contrast to the curved stochastic transport paths induced by diffusion, FlowDet learns simpler and straighter paths resulting in faster scaling of detection performance as the number of inference steps grows. We find that this reformulation enables us to outperform diffusion based detection systems (as well as non-generative baselines) across a wide range of experiments, including various precision/recall operating points using multiple feature backbones and datasets. In particular, when evaluating under recall-constrained settings, we can highlight the effects of the generative transport without over-compensating with large numbers of proposals. This provides gains of up to +3.6% AP and +4.2% AP$_{rare}$ over DiffusionDet on the COCO and LVIS datasets, respectively. --- ### GinSign: Grounding Natural Language Into System Signatures for Temporal Logic Translation Authors: William English, Chase Walker, Dominic Simon, Rickard Ewetz Published: 2025-12-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.16770v1 Natural language (NL) to temporal logic (TL) translation enables engineers to specify, verify, and enforce system behaviors without manually crafting formal specifications-an essential capability for building trustworthy autonomous systems. While existing NL-to-TL translation frameworks have demonstrated encouraging initial results, these systems either explicitly assume access to accurate atom grounding or suffer from low grounded translation accuracy. In this paper, we propose a framework for Grounding Natural Language Into System Signatures for Temporal Logic translation called GinSign. The framework introduces a grounding model that learns the abstract task of mapping NL spans onto a given system signature: given a lifted NL specification and a system signature $\mathcal{S}$, the classifier must assign each lifted atomic proposition to an element of the set of signature-defined atoms $\mathcal{P}$. We decompose the grounding task hierarchically -- first predicting predicate labels, then selecting the appropriately typed constant arguments. Decomposing this task from a free-form generation problem into a structured classification problem permits the use of smaller masked language models and eliminates the reliance on expensive LLMs. Experiments across multiple domains show that frameworks which omit grounding tend to produce syntactically correct lifted LTL that is semantically nonequivalent to grounded target expressions, whereas our framework supports downstream model checking and achieves grounded logical-equivalence scores of $95.5\%$, a $1.4\times$ improvement over SOTA. --- ### On The Hidden Biases of Flow Matching Samplers Authors: Soon Hoe Lim Published: 2025-12-18 Categories: stat.ML, cs.LG, math.PR arXiv: https://arxiv.org/abs/2512.16768v3 Flow matching (FM) constructs continuous-time ODE samplers by prescribing probability paths between a base distribution and a target distribution. In this note, we study FM through the lens of finite-sample plug-in estimation. In addition to replacing population expectations by sample averages, one may replace the target distribution itself by a finite-sample surrogate, ranging from the empirical measure to a smoothed estimator. This viewpoint yields a natural hierarchy of empirical FM models. For affine conditional flows, we derive the exact empirical minimizer and identify a smoothed plug-in regime in which the terminal law is exactly a kernel-mixture estimator. This plug-in perspective clarifies several coupled finite-sample biases of empirical FM. First, replacing the target law by a finite-sample surrogate changes the statistical target. Second, the empirical minimizer is generally not a gradient field, even when each conditional flow is. Third, a fixed empirical marginal path does not determine a unique particle dynamics: one may add extra vector fields whose probability flux has zero divergence without changing the marginal path. For Gaussian affine conditional paths, we give explicit families of such flux-null corrections. Finally, the source distribution provides a primary mechanism controlling upper tails of kinetic energy. In particular, Gaussian bases yield exponential upper-tail bounds for instantaneous and integrated kinetic energies, whereas polynomially tailed bases yield corresponding polynomial upper-tail bounds. --- ### On The Hidden Biases of Flow Matching Samplers Authors: Soon Hoe Lim Published: 2025-12-18 Categories: stat.ML, cs.LG, math.PR arXiv: https://arxiv.org/abs/2512.16768v1 We study the implicit bias of flow matching (FM) samplers via the lens of empirical flow matching. Although population FM may produce gradient-field velocities resembling optimal transport (OT), we show that the empirical FM minimizer is almost never a gradient field, even when each conditional flow is. Consequently, empirical FM is intrinsically energetically suboptimal. In view of this, we analyze the kinetic energy of generated samples. With Gaussian sources, both instantaneous and integrated kinetic energies exhibit exponential concentration, while heavy-tailed sources lead to polynomial tails. These behaviors are governed primarily by the choice of source distribution rather than the data. Overall, these notes provide a concise mathematical account of the structural and energetic biases arising in empirical FM. ---