# Research Papers — 2025-12 · bundle page 155 of 175 > Full-text markdown of papers #7701–7750 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/155 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/154.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/156.md) ### A Theoretical Framework for Auxiliary-Loss-Free Load Balancing of Sparse Mixture-of-Experts in Large-Scale AI Models Authors: X. Y. Han, Yuan Zhong Published: 2025-12-03 Categories: math.OC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.03915v3 In large-scale AI training, Sparse Mixture-of-Experts (s-MoE) layers enable scaling by activating only a small subset of experts per token. An operational challenge in this design is load balancing: routing tokens to minimize the number of idle experts, which is important for the efficient utilization of costly GPUs and for the thorough training of architecture parameters across all experts. We provide a theoretical framework for analyzing the Auxiliary-Loss-Free Load Balancing (ALF-LB) procedure -- proposed by DeepSeek's Wang et al. (2024) -- by casting it as a primal-dual method using a single-shot, constant-time update per training iteration for solving an assignment problem. First, in a stylized deterministic setting, our framework yields several insightful structural properties: (i) a monotonic improvement condition for the Lagrangian objective, (ii) a preference rule that moves tokens from overloaded to underloaded experts, and (iii) an approximate-balancing guarantee. Then, we incorporate the stochastic and dynamic nature of AI training using a generalized online optimization formulation. In the online setting, we derive a strong convexity property of the objective that leads to a logarithmic expected regret bound under certain step-size choices. Additionally, we present real experiments on 1B-parameter DeepSeekMoE models to complement our theoretical findings. Together, these results build a principled framework for analyzing the Auxiliary-Loss-Free Load Balancing of s-MoE in AI models. --- ### A Theoretical Framework for Auxiliary-Loss-Free Load Balancing of Sparse Mixture-of-Experts in Large-Scale AI Models Authors: X. Y. Han, Yuan Zhong Published: 2025-12-03 Categories: math.OC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.03915v1 In large-scale AI training, Sparse Mixture-of-Experts (s-MoE) layers enable scaling by activating only a small subset of experts per token. An operational challenge in this design is load balancing: routing tokens to minimize the number of idle experts, which is important for the efficient utilization of (costly) GPUs. We provide a theoretical framework for analyzing the Auxiliary-Loss-Free Load Balancing (ALF-LB) procedure -- proposed by DeepSeek's Wang et al. (2024) -- by casting it as a one-step-per-iteration primal-dual method for an assignment problem. First, in a stylized deterministic setting, our framework yields several insightful structural properties: (i) a monotonic improvement of a Lagrangian objective, (ii) a preference rule that moves tokens from overloaded to underloaded experts, and (iii) an approximate-balancing guarantee. Then, we incorporate the stochastic and dynamic nature of AI training using a generalized online optimization formulation. In the online setting, we derive a strong convexity property of the objective that leads to a logarithmic expected regret bound under certain step-size choices. Additionally, we present real experiments on 1B-parameter DeepSeekMoE models to complement our theoretical findings. Together, these results build a principled framework for analyzing the Auxiliary-Loss-Free Load Balancing of s-MoE in AI models. --- ### Hierarchical Vision Language Action Model Using Success and Failure Demonstrations Authors: Jeongeun Park, Jihwan Yoon, Byungwoo Jeon, Juhan Park, Jinwoo Shin, Namhoon Cho, Kyungjae Lee, Sangdoo Yun, Sungjoon Choi Published: 2025-12-03 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2512.03913v1 Prior Vision-Language-Action (VLA) models are typically trained on teleoperated successful demonstrations, while discarding numerous failed attempts that occur naturally during data collection. However, these failures encode where and how policies can be fragile, information that can be exploited to improve robustness. We address this problem by leveraging mixed-quality datasets to learn failure-aware reasoning at planning time. We introduce VINE, a hierarchical vision-language-action model that separates high-level reasoning (System 2) from low-level control (System 1) under a hierarchical reinforcement learning formalism, making failures usable as a structured learning signal rather than noisy supervision. System 2 performs feasibility-guided tree search over a 2D scene-graph abstraction: it proposes subgoal transitions, predicts success probabilities from both successes and failures, and prunes brittle branches before execution, effectively casting plan evaluation as feasibility scoring. The selected subgoal sequence is then passed to System 1, which executes low-level actions without modifying the agent's core skills. Trained entirely from offline teleoperation data, VINE integrates negative experience directly into the decision loop. Across challenging manipulation tasks, this approach consistently improves success rates and robustness, demonstrating that failure data is an essential resource for converting the broad competence of VLAs into robust execution. --- ### Autonomous Reinforcement Learning Robot Control with Intel's Loihi 2 Neuromorphic Hardware Authors: Kenneth Stewart, Roxana Leontie, Samantha Chapin, Joe Hays, Sumit Bam Shrestha, Carl Glen Henshaw Published: 2025-12-03 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.03911v1 We present an end-to-end pipeline for deploying reinforcement learning (RL) trained Artificial Neural Networks (ANNs) on neuromorphic hardware by converting them into spiking Sigma-Delta Neural Networks (SDNNs). We demonstrate that an ANN policy trained entirely in simulation can be transformed into an SDNN compatible with Intel's Loihi 2 architecture, enabling low-latency and energy-efficient inference. As a test case, we use an RL policy for controlling the Astrobee free-flying robot, similar to a previously hardware in space-validated controller. The policy, trained with Rectified Linear Units (ReLUs), is converted to an SDNN and deployed on Intel's Loihi 2, then evaluated in NVIDIA's Omniverse Isaac Lab simulation environment for closed-loop control of Astrobee's motion. We compare execution performance between GPU and Loihi 2. The results highlight the feasibility of using neuromorphic platforms for robotic control and establish a pathway toward energy-efficient, real-time neuromorphic computation in future space and terrestrial robotics applications. --- ### Zero-Shot Video Translation and Editing with Frame Spatial-Temporal Correspondence Authors: Shuai Yang, Junxin Lin, Yifan Zhou, Ziwei Liu, Chen Change Loy Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03905v1 The remarkable success in text-to-image diffusion models has motivated extensive investigation of their potential for video applications. Zero-shot techniques aim to adapt image diffusion models for videos without requiring further model training. Recent methods largely emphasize integrating inter-frame correspondence into attention mechanisms. However, the soft constraint applied to identify the valid features to attend is insufficient, which could lead to temporal inconsistency. In this paper, we present FRESCO, which integrates intra-frame correspondence with inter-frame correspondence to formulate a more robust spatial-temporal constraint. This enhancement ensures a consistent transformation of semantically similar content between frames. Our method goes beyond attention guidance to explicitly optimize features, achieving high spatial-temporal consistency with the input video, significantly enhancing the visual coherence of manipulated videos. We verify FRESCO adaptations on two zero-shot tasks of video-to-video translation and text-guided video editing. Comprehensive experiments demonstrate the effectiveness of our framework in generating high-quality, coherent videos, highlighting a significant advance over current zero-shot methods. --- ### BERnaT: Basque Encoders for Representing Natural Textual Diversity Authors: Ekhi Azurmendi, Joseba Fernandez de Landa, Jaione Bengoetxea, Maite Heredia, Julen Etxaniz, Mikel Zubillaga, Ander Soraluze, Aitor Soroa Published: 2025-12-03 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.03903v2 Language models depend on massive text corpora that are often filtered for quality, a process that can unintentionally exclude non-standard linguistic varieties, reduce model robustness and reinforce representational biases. In this paper, we argue that language models should aim to capture the full spectrum of language variation (dialectal, historical, informal, etc.) rather than relying solely on standardized text. Focusing on the Basque language, we construct new corpora combining standard, social media, and historical sources, and pre-train the BERnaT family of encoder-only models in three configurations: standard, diverse, and combined. We further propose an evaluation framework that separates Natural Language Understanding (NLU) tasks into standard and diverse subsets to assess linguistic generalization. Results show that models trained on both standard and diverse data consistently outperform those trained on standard corpora, improving performance across all task types without compromising standard benchmark accuracy. These findings highlight the importance of linguistic diversity in building inclusive, generalizable language models. --- ### BERnaT: Basque Encoders for Representing Natural Textual Diversity Authors: Ekhi Azurmendi, Joseba Fernandez de Landa, Jaione Bengoetxea, Maite Heredia, Julen Etxaniz, Mikel Zubillaga, Ander Soraluze, Aitor Soroa Published: 2025-12-03 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2512.03903v1 Language models depend on massive text corpora that are often filtered for quality, a process that can unintentionally exclude non-standard linguistic varieties, reduce model robustness and reinforce representational biases. In this paper, we argue that language models should aim to capture the full spectrum of language variation (dialectal, historical, informal, etc.) rather than relying solely on standardized text. Focusing on Basque, a morphologically rich and low-resource language, we construct new corpora combining standard, social media, and historical sources, and pre-train the BERnaT family of encoder-only models in three configurations: standard, diverse, and combined. We further propose an evaluation framework that separates Natural Language Understanding (NLU) tasks into standard and diverse subsets to assess linguistic generalization. Results show that models trained on both standard and diverse data consistently outperform those trained on standard corpora, improving performance across all task types without compromising standard benchmark accuracy. These findings highlight the importance of linguistic diversity in building inclusive, generalizable language models. --- ### Probabilistic Foundations of Fuzzy Simplicial Sets for Nonlinear Dimensionality Reduction Authors: Janis Keck, Lukas Silvester Barth, Fatemeh, Fahimi, Parvaneh Joharinad, Jürgen Jost Published: 2025-12-03 Categories: cs.LG, math.AT, stat.ML arXiv: https://arxiv.org/abs/2512.03899v1 Fuzzy simplicial sets have become an object of interest in dimensionality reduction and manifold learning, most prominently through their role in UMAP. However, their definition through tools from algebraic topology without a clear probabilistic interpretation detaches them from commonly used theoretical frameworks in those areas. In this work we introduce a framework that explains fuzzy simplicial sets as marginals of probability measures on simplicial sets. In particular, this perspective shows that the fuzzy weights of UMAP arise from a generative model that samples Vietoris-Rips filtrations at random scales, yielding cumulative distribution functions of pairwise distances. More generally, the framework connects fuzzy simplicial sets to probabilistic models on the face poset, clarifies the relation between Kullback-Leibler divergence and fuzzy cross-entropy in this setting, and recovers standard t-norms and t-conorms via Boolean operations on the underlying simplicial sets. We then show how new embedding methods may be derived from this framework and illustrate this on an example where we generalize UMAP using Čech filtrations with triplet sampling. In summary, this probabilistic viewpoint provides a unified probabilistic theoretical foundation for fuzzy simplicial sets, clarifies the role of UMAP within this framework, and enables the systematic derivation of new dimensionality reduction methods. --- ### Parameter efficient hybrid spiking-quantum convolutional neural network with surrogate gradient and quantum data-reupload Authors: Luu Trong Nhan, Luu Trung Duong, Pham Ngoc Nam, Truong Cong Thang Published: 2025-12-03 Categories: cs.NE arXiv: https://arxiv.org/abs/2512.03895v1 The rapid advancement of artificial intelligence (AI) and deep learning (DL) has catalyzed the emergence of several optimization-driven subfields, notably neuromorphic computing and quantum machine learning. Leveraging the differentiable nature of hybrid models, researchers have explored their potential to address complex problems through unified optimization strategies. One such development is the Spiking Quantum Neural Network (SQNN), which combines principles from spiking neural networks (SNNs) and quantum computing. However, existing SQNN implementations often depend on pretrained SNNs due to the non-differentiable nature of spiking activity and the limited scalability of current SNN encoders. In this work, we propose a novel architecture, Spiking-Quantum Data Re-upload Convolutional Neural Network (SQDR-CNN), that enables joint training of convolutional SNNs and quantum circuits within a single backpropagation framework. Unlike its predecessor, SQDR-CNN allow convergence to reasonable performance without the reliance of pretrained spiking encoder and subsetting datasets. We also clarified some theoretical foundations, testing new design using quantum data-reupload with different training algorithm-initialization and evaluate the performance of the proposed model under noisy simulated quantum environments. As a result, we were able to achieve 86% of the mean top-performing accuracy of the SOTA SNN baselines, yet uses only 0.5% of the smallest spiking model's parameters. Through this integration of neuromorphic and quantum paradigms, we aim to open new research directions and foster technological progress in multi-modal, learnable systems. --- ### Digital Twin-based Control Co-Design of Full Vehicle Active Suspensions via Deep Reinforcement Learning Authors: Ying-Kuan Tsai, Yi-Ping Chen, Vispi Karkaria, Wei Chen Published: 2025-12-03 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2512.03891v1 Active suspension systems are critical for enhancing vehicle comfort, safety, and stability, yet their performance is often limited by fixed hardware designs and control strategies that cannot adapt to uncertain and dynamic operating conditions. Recent advances in digital twins (DTs) and deep reinforcement learning (DRL) offer new opportunities for real-time, data-driven optimization across a vehicle's lifecycle. However, integrating these technologies into a unified framework remains an open challenge. This work presents a DT-based control co-design (CCD) framework for full-vehicle active suspensions using multi-generation design concepts. By integrating automatic differentiation into DRL, we jointly optimize physical suspension components and control policies under varying driver behaviors and environmental uncertainties. DRL also addresses the challenge of partial observability, where only limited states can be sensed and fed back to the controller, by learning optimal control actions directly from available sensor information. The framework incorporates model updating with quantile learning to capture data uncertainty, enabling real-time decision-making and adaptive learning from digital-physical interactions. The approach demonstrates personalized optimization of suspension systems under two distinct driving settings (mild and aggressive). Results show that the optimized systems achieve smoother trajectories and reduce control efforts by approximately 43% and 52% for mild and aggressive, respectively, while maintaining ride comfort and stability. Contributions include: developing a DT-enabled CCD framework integrating DRL and uncertainty-aware model updating for full-vehicle active suspensions, introducing a multi-generation design strategy for self-improving systems, and demonstrating personalized optimization of active suspension systems for distinct driver types. --- ### A Hierarchical Tree-based approach for creating Configurable and Static Deep Research Agent (Static-DRA) Authors: Saurav Prateek Published: 2025-12-03 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.03887v2 The advancement in Large Language Models has driven the creation of complex agentic systems, such as Deep Research Agents (DRAs), to overcome the limitations of static Retrieval Augmented Generation (RAG) pipelines in handling complex, multi-turn research tasks. This paper introduces the Static Deep Research Agent (Static-DRA), a novel solution built upon a configurable and hierarchical Tree-based static workflow. The core contribution is the integration of two user-tunable parameters, Depth and Breadth, which provide granular control over the research intensity. This design allows end-users to consciously balance the desired quality and comprehensiveness of the research report against the associated computational cost of Large Language Model (LLM) interactions. The agent's architecture, comprising Supervisor, Independent, and Worker agents, facilitates effective multi-hop information retrieval and parallel sub-topic investigation. We evaluate the Static-DRA against the established DeepResearch Bench using the RACE (Reference-based Adaptive Criteria-driven Evaluation) framework. Configured with a depth of 2 and a breadth of 5, and powered by the gemini-2.5-pro model, the agent achieved an overall score of 34.72. Our experiments validate that increasing the configured Depth and Breadth parameters results in a more in-depth research process and a correspondingly higher evaluation score. The Static-DRA offers a pragmatic and resource-aware solution, empowering users with transparent control over the deep research process. The entire source code, outputs and benchmark results are open-sourced at https://github.com/SauravP97/Static-Deep-Research/ --- ### A Hierarchical Tree-based approach for creating Configurable and Static Deep Research Agent (Static-DRA) Authors: Saurav Prateek Published: 2025-12-03 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.03887v1 The advancement in Large Language Models has driven the creation of complex agentic systems, such as Deep Research Agents (DRAs), to overcome the limitations of static Retrieval Augmented Generation (RAG) pipelines in handling complex, multi-turn research tasks. This paper introduces the Static Deep Research Agent (Static-DRA), a novel solution built upon a configurable and hierarchical Tree-based static workflow. The core contribution is the integration of two user-tunable parameters, Depth and Breadth, which provide granular control over the research intensity. This design allows end-users to consciously balance the desired quality and comprehensiveness of the research report against the associated computational cost of Large Language Model (LLM) interactions. The agent's architecture, comprising Supervisor, Independent, and Worker agents, facilitates effective multi-hop information retrieval and parallel sub-topic investigation. We evaluate the Static-DRA against the established DeepResearch Bench using the RACE (Reference-based Adaptive Criteria-driven Evaluation) framework. Configured with a depth of 2 and a breadth of 5, and powered by the gemini-2.5-pro model, the agent achieved an overall score of 34.72. Our experiments validate that increasing the configured Depth and Breadth parameters results in a more in-depth research process and a correspondingly higher evaluation score. The Static-DRA offers a pragmatic and resource-aware solution, empowering users with transparent control over the deep research process. The entire source code, outputs and benchmark results are open-sourced at https://github.com/SauravP97/Static-Deep-Research/ --- ### Dual Cross-Attention Siamese Transformer for Rectal Tumor Regrowth Assessment in Watch-and-Wait Endoscopy Authors: Jorge Tapias Gomez, Despoina Kanata, Aneesh Rangnekar, Christina Lee, Julio Garcia-Aguilar, Joshua Jesse Smith, Harini Veeraraghavan Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03883v3 Increasing evidence supports watch-and-wait (WW) surveillance for patients with rectal cancer who show clinical complete response (cCR) at restaging following total neoadjuvant treatment (TNT). However, accurate methods to early detect local regrowth (LR) from follow-up endoscopy images during WW are essential to manage care and prevent distant metastases. Hence, we developed a Siamese Swin Transformer with Dual Cross-Attention (SSDCA) to combine longitudinal endoscopic images at restaging and follow-up and distinguish cCR from LR. SSDCA leverages pretrained Swin Transformers to extract domain agnostic features and enhance robustness to imaging variations. Dual cross attention is implemented to emphasize features from the paired scans without requiring any spatial alignment to predict response. SSDCA as well as Swin-based baselines were trained using image pairs from 135 patients and evaluated on a held-out set of image pairs from 62 patients. SSDCA produced the best balanced accuracy (81.76% $\pm$ 0.04), sensitivity (90.07% $\pm$ 0.08), and specificity (72.86% $\pm$ 0.05). Robustness analysis showed stable performance irrespective of artifacts including blood, stool, telangiectasia, and poor image quality. UMAP clustering of extracted features showed maximal inter-cluster separation (1.45 $\pm$ 0.18) and minimal intra-cluster dispersion (1.07 $\pm$ 0.19) with SSDCA, confirming discriminative representation learning. Code and weights available at: https://github.com/Jotanator/SSDCA --- ### Dual Cross-Attention Siamese Transformer for Rectal Tumor Regrowth Assessment in Watch-and-Wait Endoscopy Authors: Jorge Tapias Gomez, Despoina Kanata, Aneesh Rangnekar, Christina Lee, Julio Garcia-Aguilar, Joshua Jesse Smith, Harini Veeraraghavan Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03883v1 Increasing evidence supports watch-and-wait (WW) surveillance for patients with rectal cancer who show clinical complete response (cCR) at restaging following total neoadjuvant treatment (TNT). However, objectively accurate methods to early detect local regrowth (LR) from follow-up endoscopy images during WW are essential to manage care and prevent distant metastases. Hence, we developed a Siamese Swin Transformer with Dual Cross-Attention (SSDCA) to combine longitudinal endoscopic images at restaging and follow-up and distinguish cCR from LR. SSDCA leverages pretrained Swin transformers to extract domain agnostic features and enhance robustness to imaging variations. Dual cross attention is implemented to emphasize features from the two scans without requiring any spatial alignment of images to predict response. SSDCA as well as Swin-based baselines were trained using image pairs from 135 patients and evaluated on a held-out set of image pairs from 62 patients. SSDCA produced the best balanced accuracy (81.76\% $\pm$ 0.04), sensitivity (90.07\% $\pm$ 0.08), and specificity (72.86\% $\pm$ 0.05). Robustness analysis showed stable performance irrespective of artifacts including blood, stool, telangiectasia, and poor image quality. UMAP clustering of extracted features showed maximal inter-cluster separation (1.45 $\pm$ 0.18) and minimal intra-cluster dispersion (1.07 $\pm$ 0.19) with SSDCA, confirming discriminative representation learning. --- ### Automatic Attack Discovery for Few-Shot Class-Incremental Learning via Large Language Models Authors: Haidong Kang, Wei Wu, Hanling Wang Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03882v1 Few-shot class incremental learning (FSCIL) is a more realistic and challenging paradigm in continual learning to incrementally learn unseen classes and overcome catastrophic forgetting on base classes with only a few training examples. Previous efforts have primarily centered around studying more effective FSCIL approaches. By contrast, less attention was devoted to thinking the security issues in contributing to FSCIL. This paper aims to provide a holistic study of the impact of attacks on FSCIL. We first derive insights by systematically exploring how human expert-designed attack methods (i.e., PGD, FGSM) affect FSCIL. We find that those methods either fail to attack base classes, or suffer from huge labor costs due to relying on huge expert knowledge. This highlights the need to craft a specialized attack method for FSCIL. Grounded in these insights, in this paper, we propose a simple yet effective ACraft method to automatically steer and discover optimal attack methods targeted at FSCIL by leveraging Large Language Models (LLMs) without human experts. Moreover, to improve the reasoning between LLMs and FSCIL, we introduce a novel Proximal Policy Optimization (PPO) based reinforcement learning to optimize learning, making LLMs generate better attack methods in the next generation by establishing positive feedback. Experiments on mainstream benchmarks show that our ACraft significantly degrades the performance of state-of-the-art FSCIL methods and dramatically beyond human expert-designed attack methods while maintaining the lowest costs of attack. --- ### Hybrid Temporal-8-Bit Spike Coding for Spiking Neural Network Surrogate Training Authors: Luu Trong Nhan, Luu Trung Duong, Pham Ngoc Nam, Truong Cong Thang Published: 2025-12-03 Categories: cs.NE arXiv: https://arxiv.org/abs/2512.03879v1 Spiking neural networks (SNNs) have emerged as a promising direction in both computational neuroscience and artificial intelligence, offering advantages such as strong biological plausibility and low energy consumption on neuromorphic hardware. Despite these benefits, SNNs still face challenges in achieving state-of-the-art performance on vision tasks. Recent work has shown that hybrid rate-temporal coding strategies (particularly those incorporating bit-plane representations of images into traditional rate coding schemes) can significantly improve performance when trained with surrogate backpropagation. Motivated by these findings, this study proposes a hybrid temporal-bit spike coding method that integrates bit-plane decompositions with temporal coding principles. Through extensive experiments across multiple computer vision benchmarks, we demonstrate that blending bit-plane information with temporal coding yields competitive, and in some cases improved, performance compared to established spike-coding techniques. To the best of our knowledge, this is the first work to introduce a hybrid temporal-bit coding scheme specifically designed for surrogate gradient training of SNNs. --- ### OmniDexVLG: Learning Dexterous Grasp Generation from Vision Language Model-Guided Grasp Semantics, Taxonomy and Functional Affordance Authors: Lei Zhang, Diwen Zheng, Kaixin Bai, Zhenshan Bing, Zoltan-Csaba Marton, Zhaopeng Chen, Alois Christian Knoll, Jianwei Zhang Published: 2025-12-03 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2512.03874v1 Dexterous grasp generation aims to produce grasp poses that align with task requirements and human interpretable grasp semantics. However, achieving semantically controllable dexterous grasp synthesis remains highly challenging due to the lack of unified modeling of multiple semantic dimensions, including grasp taxonomy, contact semantics, and functional affordance. To address these limitations, we present OmniDexVLG, a multimodal, semantics aware grasp generation framework capable of producing structurally diverse and semantically coherent dexterous grasps under joint language and visual guidance. Our approach begins with OmniDexDataGen, a semantic rich dexterous grasp dataset generation pipeline that integrates grasp taxonomy guided configuration sampling, functional affordance contact point sampling, taxonomy aware differential force closure grasp sampling, and physics based optimization and validation, enabling systematic coverage of diverse grasp types. We further introduce OmniDexReasoner, a multimodal grasp type semantic reasoning module that leverages multi agent collaboration, retrieval augmented generation, and chain of thought reasoning to infer grasp related semantics and generate high quality annotations that align language instructions with task specific grasp intent. Building upon these components, we develop a unified Vision Language Grasping generation model that explicitly incorporates grasp taxonomy, contact structure, and functional affordance semantics, enabling fine grained control over grasp synthesis from natural language instructions. Extensive experiments in simulation and real world object grasping and ablation studies demonstrate that our method substantially outperforms state of the art approaches in terms of grasp diversity, contact semantic diversity, functional affordance diversity, and semantic consistency. --- ### Reconstructing KV Caches with Cross-layer Fusion For Enhanced Transformers Authors: Hongzhan Lin, Zhiqi Bai, Xinmiao Zhang, Sen Yang, Xiang Li, Siran Yang, Yunlong Xu, Jiaheng Liu, Yongchi Zhao, Jiamang Wang, Yuchi Xu, Wenbo Su et al. Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03870v3 Transformer decoders have achieved strong results across tasks, but the memory required for the KV cache becomes prohibitive at long sequence lengths. Although Cross-layer KV Cache sharing (e.g., YOCO, CLA) offers a path to mitigate KV Cache bottleneck, it typically underperforms within-layer methods like GQA. To understand the root cause, we investigate the information flow of keys and values of the top-layers. Our preliminary reveals a clear distribution: values are predominantly derived from the bottom layer, while keys draw more information from both bottom and middle layers. Building upon this, we propose FusedKV, whose top-layer KV caches are a learnable fusion of the most informative ones from the bottom and middle layers. This fusion operates directly on post-RoPE keys, preserving relative positional information without the computational cost of re-applying rotary embeddings. To further improve efficiency, we propose FusedKV-Lite, an cross-layer sharing approach, where top-layer KV caches are directly derived from the bottom-layer values and the middle-layer keys. Compared to FusedKV, FusedKV-Lite reduces I/O overhead at the cost of a slight increase in perplexity. In experiments on LLMs ranging from 332M to 4B parameters, our proposed method reduce 50\% cache memory while achieving lower validation perplexity than the standard Transformer decoder, establishing it as a memory-efficient, high-performance architectural alternative. --- ### Reconstructing KV Caches with Cross-layer Fusion For Enhanced Transformers Authors: Hongzhan Lin, Zhiqi Bai, Xinmiao Zhang, Sen Yang, Xiang Li, Siran Yang, Yunlong Xu, Jiaheng Liu, Yongchi Zhao, Jiamang Wang, Yuchi Xu, Wenbo Su et al. Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03870v1 Transformer decoders have achieved strong results across tasks, but the memory required for the KV cache becomes prohibitive at long sequence lengths. Although Cross-layer KV Cache sharing (e.g., YOCO, CLA) offers a path to mitigate KV Cache bottleneck, it typically underperforms within-layer methods like GQA. To understand the root cause, we investigate the information flow of keys and values of the top-layers. Our preliminary reveals a clear distribution: values are predominantly derived from the bottom layer, while keys draw more information from both bottom and middle layers. Building upon this, we propose FusedKV, whose top-layer KV caches are a learnable fusion of the most informative ones from the bottom and middle layers. This fusion operates directly on post-RoPE keys, preserving relative positional information without the computational cost of re-applying rotary embeddings. To further improve efficiency, we propose FusedKV-Lite, an cross-layer sharing approach, where top-layer KV caches are directly derived from the bottom-layer values and the middle-layer keys. Compared to FusedKV, FusedKV-Lite reduces I/O overhead at the cost of a slight increase in perplexity. In experiments on LLMs ranging from 332M to 4B parameters, our proposed method reduce 50\% cache memory while achieving lower validation perplexity than the standard Transformer decoder, establishing it as a memory-efficient, high-performance architectural alternative. --- ### An Automated Framework for Large-Scale Graph-Based Cerebrovascular Analysis Authors: Daniele Falcetta, Liane S. Canas, Lorenzo Suppa, Matteo Pentassuglia, Jon Cleary, Marc Modat, Sébastien Ourselin, Maria A. Zuluaga Published: 2025-12-03 Categories: cs.CV, cs.CY arXiv: https://arxiv.org/abs/2512.03869v4 We present CaravelMetrics, a computational framework for automated cerebrovascular analysis that models vessel morphology through skeletonization-derived graph representations. The framework integrates atlas-based regional parcellation, centerline extraction, and graph construction to compute fifteen morphometric, topological, fractal, and geometric features. The features can be estimated globally from the complete vascular network or regionally within arterial territories, enabling multiscale characterization of cerebrovascular organization. Applied to 570 3D TOF-MRA scans from the IXI dataset (ages 20-86), CaravelMetrics yields reproducible vessel graphs capturing age- and sex-related variations and education-associated increases in vascular complexity, consistent with findings reported in the literature. The framework provides a scalable and fully automated approach for quantitative cerebrovascular feature extraction, supporting normative modeling and population-level studies of vascular health and aging. --- ### An Automated Framework for Large-Scale Graph-Based Cerebrovascular Analysis Authors: Daniele Falcetta, Liane S. Canas, Lorenzo Suppa, Matteo Pentassuglia, Jon Cleary, Marc Modat, Sébastien Ourselin, Maria A. Zuluaga Published: 2025-12-03 Categories: cs.CV, cs.CY arXiv: https://arxiv.org/abs/2512.03869v1 We present CaravelMetrics, a computational framework for automated cerebrovascular analysis that models vessel morphology through skeletonization-derived graph representations. The framework integrates atlas-based regional parcellation, centerline extraction, and graph construction to compute fifteen morphometric, topological, fractal, and geometric features. The features can be estimated globally from the complete vascular network or regionally within arterial territories, enabling multiscale characterization of cerebrovascular organization. Applied to 570 3D TOF-MRA scans from the IXI dataset (ages 20-86), CaravelMetrics yields reproducible vessel graphs capturing age- and sex-related variations and education-associated increases in vascular complexity, consistent with findings reported in the literature. The framework provides a scalable and fully automated approach for quantitative cerebrovascular feature extraction, supporting normative modeling and population-level studies of vascular health and aging. --- ### Hyperdimensional Computing for Sustainable Manufacturing: An Initial Assessment Authors: Danny Hoang, Anandkumar Patel, Ruimen Chen, Rajiv Malhotra, Farhad Imani Published: 2025-12-03 Categories: cs.LG, cs.AI, cs.PF, cs.SC arXiv: https://arxiv.org/abs/2512.03864v1 Smart manufacturing can significantly improve efficiency and reduce energy consumption, yet the energy demands of AI models may offset these gains. This study utilizes in-situ sensing-based prediction of geometric quality in smart machining to compare the energy consumption, accuracy, and speed of common AI models. HyperDimensional Computing (HDC) is introduced as an alternative, achieving accuracy comparable to conventional models while drastically reducing energy consumption, 200$\times$ for training and 175 to 1000$\times$ for inference. Furthermore, HDC reduces training times by 200$\times$ and inference times by 300 to 600$\times$, showcasing its potential for energy-efficient smart manufacturing. --- ### Diminishing Returns in Self-Supervised Learning Authors: Oli Bridge, Huey Sun, Botond Branyicskai-Nagy, Charles D'Ornano, Shomit Basu Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03862v2 Transformer-based architectures have become a dominant paradigm in vision and language, but their success is often attributed to large model capacity and massive training data. In this work, we examine how self-supervised pre-training, intermediate fine-tuning, and downstream fine-tuning interact in a low-capacity regime, using a 5M-parameter Vision Transformer for semantic segmentation. Across multiple data scales, we find that masked image modeling pre-training and downstream fine-tuning reliably improve performance, but with clear diminishing returns as supervision increases. In contrast, inserting an intermediate classification fine-tuning stage consistently degrades downstream performance, with the largest drops occurring precisely where pre-training is most effective. Through an analysis of patch-level representation geometry, we show that classification-based intermediate supervision actively interferes with representations learned during pre-training by collapsing spatial structure critical for dense prediction. These results indicate that, in small models, the geometry of supervision matters more than the number of training stages: misaligned intermediate objectives can negate the benefits of pre-training rather than amplify them. --- ### Diminishing Returns in Self-Supervised Learning Authors: Oli Bridge, Huey Sun, Botond Branyicskai-Nagy, Charles D'Ornano, Shomit Basu Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03862v1 While transformer-based architectures have taken computer vision and NLP by storm, they often require a vast amount of parameters and training data to attain strong performance. In this work, we experiment with three distinct pre-training, intermediate fine-tuning, and downstream datasets and training objectives to explore their marginal benefits on a small 5M-parameter vision transformer. We find that while pre-training and fine-tuning always help our model but have diminishing returns, intermediate fine-tuning can actually show harmful impact on downstream performance, potentially due to dissimilarity in task mechanics. Taken together, our results suggest that small-scale ViTs benefit most from targeted pre-training and careful data selection, while indiscriminate stacking of intermediate tasks can waste compute and even degrade performance. --- ### Scalable Decision Focused Learning via Online Trainable Surrogates Authors: Gaetano Signorelli, Michele Lombardi Published: 2025-12-03 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.03861v1 Decision support systems often rely on solving complex optimization problems that may require to estimate uncertain parameters beforehand. Recent studies have shown how using traditionally trained estimators for this task can lead to suboptimal solutions. Using the actual decision cost as a loss function (called Decision Focused Learning) can address this issue, but with a severe loss of scalability at training time. To address this issue, we propose an acceleration method based on replacing costly loss function evaluations with an efficient surrogate. Unlike previously defined surrogates, our approach relies on unbiased estimators reducing the risk of spurious local optima and can provide information on its local confidence allowing one to switch to a fallback method when needed. Furthermore, the surrogate is designed for a black-box setting, which enables compensating for simplifications in the optimization model and accounting for recourse actions during cost computation. In our results, the method reduces costly inner solver calls, with a solution quality comparable to other state-of-the-art techniques. --- ### The PAR dataset: Prostate biopsy whole slide images from an underrepresented Middle Eastern population Authors: Peshawa J. Muhammad Ali, Navin Vincent, Saman S. Abdulla, Han N. Mohammed Fadhl, Anders Blilie, Kelvin Szolnoky, Julia Anna Mielcarz, Xiaoyi Ji, Kimmo Kartasalo, Abdulbasit K. Al-Talabani, Nita Mulliqi Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03854v2 Artificial intelligence (AI) is increasingly used in digital pathology. Publicly available histopathology datasets remain scarce, and those that do exist predominantly represent Western populations. Consequently, the generalizability of AI models to populations from less digitized regions, such as the Middle East, is largely unknown. This motivates the public release of our dataset to support the development and validation of pathology AI models across globally diverse populations. We present 1,017 whole slide images by digitizing 339 glass slides of prostate core needle biopsies from a consecutive series of 185 patients collected in Erbil, Iraq. The slides are associated with Gleason scores and International Society of Urological Pathology grades assigned independently by three pathologists. Scanning was performed using two high-throughput scanners (Leica and Hamamatsu) and one compact scanner (Grundium). All slides were de-identified and are provided in their native formats without further conversion. The dataset enables grading concordance analyses, color normalization, and cross-scanner robustness evaluations. The PAR dataset has been deposited to the BioImage Archive under accession number S-BIAD2323 (https:// doi.org/10.6019/S-BIAD2323). --- ### Prostate biopsy whole slide image dataset from an underrepresented Middle Eastern population Authors: Peshawa J. Muhammad Ali, Navin Vincent, Saman S. Abdulla, Han N. Mohammed Fadhl, Anders Blilie, Kelvin Szolnoky, Julia Anna Mielcarz, Xiaoyi Ji, Kimmo Kartasalo, Abdulbasit K. Al-Talabani, Nita Mulliqi Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03854v1 Artificial intelligence (AI) is increasingly used in digital pathology. Publicly available histopathology datasets remain scarce, and those that do exist predominantly represent Western populations. Consequently, the generalizability of AI models to populations from less digitized regions, such as the Middle East, is largely unknown. This motivates the public release of our dataset to support the development and validation of pathology AI models across globally diverse populations. We present 339 whole-slide images of prostate core needle biopsies from a consecutive series of 185 patients collected in Erbil, Iraq. The slides are associated with Gleason scores and International Society of Urological Pathology grades assigned independently by three pathologists. Scanning was performed using two high-throughput scanners (Leica and Hamamatsu) and one compact scanner (Grundium). All slides were de-identified and are provided in their native formats without further conversion. The dataset enables grading concordance analyses, color normalization, and cross-scanner robustness evaluations. Data will be deposited in the Bioimage Archive (BIA) under accession code: to be announced (TBA), and released under a CC BY 4.0 license. --- ### Traffic Image Restoration under Adverse Weather via Frequency-Aware Mamba Authors: Liwen Pan, Longguang Wang, Guangwei Gao, Jun Wang, Jun Shi, Juncheng Li Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03852v1 Traffic image restoration under adverse weather conditions remains a critical challenge for intelligent transportation systems. Existing methods primarily focus on spatial-domain modeling but neglect frequency-domain priors. Although the emerging Mamba architecture excels at long-range dependency modeling through patch-wise correlation analysis, its potential for frequency-domain feature extraction remains unexplored. To address this, we propose Frequency-Aware Mamba (FAMamba), a novel framework that integrates frequency guidance with sequence modeling for efficient image restoration. Our architecture consists of two key components: (1) a Dual-Branch Feature Extraction Block (DFEB) that enhances local-global interaction via bidirectional 2D frequency-adaptive scanning, dynamically adjusting traversal paths based on sub-band texture distributions; and (2) a Prior-Guided Block (PGB) that refines texture details through wavelet-based high-frequency residual learning, enabling high-quality image reconstruction with precise details. Meanwhile, we design a novel Adaptive Frequency Scanning Mechanism (AFSM) for the Mamba architecture, which enables the Mamba to achieve frequency-domain scanning across distinct subgraphs, thereby fully leveraging the texture distribution characteristics inherent in subgraph structures. Extensive experiments demonstrate the efficiency and effectiveness of FAMamba. --- ### Comparison of neural network training strategies for the simulation of dynamical systems Authors: Paul Strasser, Andreas Pfeffer, Jakob Weber, Markus Gurtner, Andreas Körner Published: 2025-12-03 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2512.03851v2 Neural networks have become a widely adopted tool for modeling nonlinear dynamical systems from data. However, the choice of training strategy remains a key design decision, particularly for simulation tasks. This paper compares two predominant strategies: parallel and series-parallel training. The conducted empirical analysis spans five neural network architectures and two examples: a pneumatic valve test bench and an industrial robot benchmark. The study reveals that, even though series-parallel training dominates current practice, parallel training consistently yields better long-term prediction accuracy. Additionally, this work clarifies the often inconsistent terminology in the literature and relate both strategies to concepts from system identification. The findings suggest that parallel training should be considered the default training strategy for neural network-based simulation of dynamical systems. --- ### Comparison of neural network training strategies for the simulation of dynamical systems Authors: Paul Strasser, Andreas Pfeffer, Jakob Weber, Markus Gurtner, Andreas Körner Published: 2025-12-03 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2512.03851v1 Neural networks have become a widely adopted tool for modeling nonlinear dynamical systems from data. However, the choice of training strategy remains a key design decision, particularly for simulation tasks. This paper compares two predominant strategies: parallel and series-parallel training. The conducted empirical analysis spans five neural network architectures and two examples: a pneumatic valve test bench and an industrial robot benchmark. The study reveals that, even though series-parallel training dominates current practice, parallel training consistently yields better long-term prediction accuracy. Additionally, this work clarifies the often inconsistent terminology in the literature and relate both strategies to concepts from system identification. The findings suggest that parallel training should be considered the default training strategy for neural network-based simulation of dynamical systems. --- ### PULSE: A Unified Multi-Task Architecture for Cardiac Segmentation, Diagnosis, and Few-Shot Cross-Modality Clinical Adaptation Authors: Hania Ghouse, Maryam Alsharqi, Farhad R. Nezami, Muzammil Behzad Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03848v1 Cardiac image analysis remains fragmented across tasks: anatomical segmentation, disease classification, and grounded clinical report generation are typically handled by separate networks trained under different data regimes. No existing framework unifies these objectives within a single architecture while retaining generalization across imaging modalities and datasets. We introduce PULSE, a multi-task vision-language framework built on self-supervised representations and optimized through a composite supervision strategy that balances region overlap learning, pixel wise classification fidelity, and boundary aware IoU refinement. A multi-scale token reconstruction decoder enables anatomical segmentation, while shared global representations support disease classification and clinically grounded text output allowing the model to transition from pixels to structures and finally clinical reasoning within one architecture. Unlike prior task-specific pipelines, PULSE learns task-invariant cardiac priors, generalizes robustly across datasets, and can be adapted to new imaging modalities with minimal supervision. This moves the field closer to a scalable, foundation style cardiac analysis framework. --- ### DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training Authors: Dingwei Zhu, Zhiheng Xi, Shihan Dou, Yuhui Wang, Sixian Li, Junjie Ye, Honglin Guo, Shichun Liu, Chenhao Huang, Yajie Yang, Junlin Shang, Senjie Jin et al. Published: 2025-12-03 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.03847v3 Reinforcement learning (RL) has shown strong performance in LLM post-training, but real-world deployment often involves noisy or incomplete supervision. In such settings, complex and unreliable supervision signals can destabilize training and harm generalization. While existing approaches such as worst-case optimization (e.g., RFQI, CQL) and mean-based methods (e.g., PPO, GRPO) can improve stability, they often overlook generalization and may produce overly conservative policies, leading to uneven performance across diverse real scenarios. To this end, we introduce DVPO (Distributional Value Modeling with Risk-aware Policy Optimization), a new RL framework that combines conditional risk theory with distributional value modeling to better balance robustness and generalization. DVPO learns token-level value distributions to provide fine-grained supervision, and applies an asymmetric risk regularization to shape the distribution tails: it contracts the lower tail to dampen noisy negative deviations, while expanding the upper tail to preserve exploratory diversity. Across extensive experiments and analysis in multi-turn dialogue, math reasoning, and scientific QA, DVPO consistently outperforms PPO, GRPO, and robust Bellman-based PPO under noisy supervision, showing its potential for LLM post-training in the real-world. --- ### DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training Authors: Dingwei Zhu, Zhiheng Xi, Shihan Dou, Yuhui Wang, Sixian Li, Junjie Ye, Honglin Guo, Shichun Liu, Chenhao Huang, Yajie Yang, Junlin Shang, Senjie Jin et al. Published: 2025-12-03 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.03847v1 Reinforcement learning (RL) has shown strong performance in LLM post-training, but real-world deployment often involves noisy or incomplete supervision. In such settings, complex and unreliable supervision signals can destabilize training and harm generalization. While existing approaches such as worst-case optimization (e.g., RFQI, CQL) and mean-based methods (e.g., PPO, GRPO) can improve stability, they often overlook generalization and may produce overly conservative policies, leading to uneven performance across diverse real scenarios. To this end, we introduce DVPO (Distributional Value Modeling with Risk-aware Policy Optimization), a new RL framework that combines conditional risk theory with distributional value modeling to better balance robustness and generalization. DVPO learns token-level value distributions to provide fine-grained supervision, and applies an asymmetric risk regularization to shape the distribution tails: it contracts the lower tail to dampen noisy negative deviations, while expanding the upper tail to preserve exploratory diversity. Across extensive experiments and analysis in multi-turn dialogue, math reasoning, and scientific QA, DVPO consistently outperforms PPO, GRPO, and robust Bellman-based PPO under noisy supervision, showing its potential for LLM post-training in the real-world. --- ### CoDA: From Text-to-Image Diffusion Models to Training-Free Dataset Distillation Authors: Letian Zhou, Songhua Liu, Xinchao Wang Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03844v1 Prevailing Dataset Distillation (DD) methods leveraging generative models confront two fundamental limitations. First, despite pioneering the use of diffusion models in DD and delivering impressive performance, the vast majority of approaches paradoxically require a diffusion model pre-trained on the full target dataset, undermining the very purpose of DD and incurring prohibitive training costs. Second, although some methods turn to general text-to-image models without relying on such target-specific training, they suffer from a significant distributional mismatch, as the web-scale priors encapsulated in these foundation models fail to faithfully capture the target-specific semantics, leading to suboptimal performance. To tackle these challenges, we propose Core Distribution Alignment (CoDA), a framework that enables effective DD using only an off-the-shelf text-to-image model. Our key idea is to first identify the "intrinsic core distribution" of the target dataset using a robust density-based discovery mechanism. We then steer the generative process to align the generated samples with this core distribution. By doing so, CoDA effectively bridges the gap between general-purpose generative priors and target semantics, yielding highly representative distilled datasets. Extensive experiments suggest that, without relying on a generative model specifically trained on the target dataset, CoDA achieves performance on par with or even superior to previous methods with such reliance across all benchmarks, including ImageNet-1K and its subsets. Notably, it establishes a new state-of-the-art accuracy of 60.4% at the 50-images-per-class (IPC) setup on ImageNet-1K. Our code is available on the project webpage: https://github.com/zzzlt422/CoDA --- ### Training and Evaluation of Guideline-Based Medical Reasoning in LLMs Authors: Michael Staniek, Artem Sokolov, Stefan Riezler Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03838v1 Machine learning for early prediction in medicine has recently shown breakthrough performance, however, the focus on improving prediction accuracy has led to a neglect of faithful explanations that are required to gain the trust of medical practitioners. The goal of this paper is to teach LLMs to follow medical consensus guidelines step-by-step in their reasoning and prediction process. Since consensus guidelines are ubiquitous in medicine, instantiations of verbalized medical inference rules to electronic health records provide data for fine-tuning LLMs to learn consensus rules and possible exceptions thereof for many medical areas. Consensus rules also enable an automatic evaluation of the model's inference process regarding its derivation correctness (evaluating correct and faithful deduction of a conclusion from given premises) and value correctness (comparing predicted values against real-world measurements). We exemplify our work using the complex Sepsis-3 consensus definition. Our experiments show that small fine-tuned models outperform one-shot learning of considerably larger LLMs that are prompted with the explicit definition and models that are trained on medical texts including consensus definitions. Since fine-tuning on verbalized rule instantiations of a specific medical area yields nearly perfect derivation correctness for rules (and exceptions) on unseen patient data in that area, the bottleneck for early prediction is not out-of-distribution generalization, but the orthogonal problem of generalization into the future by forecasting sparsely and irregularly sampled clinical variables. We show that the latter results can be improved by integrating the output representations of a time series forecasting model with the LLM in a multimodal setup. --- ### Heatmap Pooling Network for Action Recognition from RGB Videos Authors: Mengyuan Liu, Jinfu Liu, Yongkang Jiang, Bin He Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03837v1 Human action recognition (HAR) in videos has garnered widespread attention due to the rich information in RGB videos. Nevertheless, existing methods for extracting deep features from RGB videos face challenges such as information redundancy, susceptibility to noise and high storage costs. To address these issues and fully harness the useful information in videos, we propose a novel heatmap pooling network (HP-Net) for action recognition from videos, which extracts information-rich, robust and concise pooled features of the human body in videos through a feedback pooling module. The extracted pooled features demonstrate obvious performance advantages over the previously obtained pose data and heatmap features from videos. In addition, we design a spatial-motion co-learning module and a text refinement modulation module to integrate the extracted pooled features with other multimodal data, enabling more robust action recognition. Extensive experiments on several benchmarks namely NTU RGB+D 60, NTU RGB+D 120, Toyota-Smarthome and UAV-Human consistently verify the effectiveness of our HP-Net, which outperforms the existing human action recognition methods. Our code is publicly available at: https://github.com/liujf69/HPNet-Action. --- ### Lean Unet: A Compact Model for Image Segmentation Authors: Ture Hassler, Ida Åkerholm, Marcus Nordström, Gabriele Balletti, Orcun Goksel Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03834v1 Unet and its variations have been standard in semantic image segmentation, especially for computer assisted radiology. Current Unet architectures iteratively downsample spatial resolution while increasing channel dimensions to preserve information content. Such a structure demands a large memory footprint, limiting training batch sizes and increasing inference latency. Channel pruning compresses Unet architecture without accuracy loss, but requires lengthy optimization and may not generalize across tasks and datasets. By investigating Unet pruning, we hypothesize that the final structure is the crucial factor, not the channel selection strategy of pruning. Based on our observations, we propose a lean Unet architecture (LUnet) with a compact, flat hierarchy where channels are not doubled as resolution is halved. We evaluate on a public MRI dataset allowing comparable reporting, as well as on two internal CT datasets. We show that a state-of-the-art pruning solution (STAMP) mainly prunes from the layers with the highest number of channels. Comparatively, simply eliminating a random channel at the pruning-identified layer or at the largest layer achieves similar or better performance. Our proposed LUnet with fixed architectures and over 30 times fewer parameters achieves performance comparable to both conventional Unet counterparts and data-adaptively pruned networks. The proposed lean Unet with constant channel count across layers requires far fewer parameters while achieving performance superior to standard Unet for the same total number of parameters. Skip connections allow Unet bottleneck channels to be largely reduced, unlike standard encoder-decoder architectures requiring increased bottleneck channels for information propagation. --- ### A Robust Camera-based Method for Breath Rate Measurement Authors: Alexey Protopopov Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03827v1 Proliferation of cheap and accessible cameras makes it possible to measure a subject's breath rate from video footage alone. Recent works on this topic have proposed a variety of approaches for accurately measuring human breath rate, however they are either tested in near-ideal conditions, or produce results that are not sufficiently accurate. The present study proposes a more robust method to measure breath rate in humans with minimal hardware requirements using a combination of mathematical transforms with a relative deviation from the ground truth of less than 5%. The method was tested on videos taken from 14 volunteers with a total duration of over 2 hours 30 minutes. The obtained results were compared to reference data and the average mean absolute error was found to be at 0.57 respirations per minute, which is noticeably better than the results from previous works. The breath rate measurement method proposed in the present article is more resistant to distortions caused by subject movement and thus allows one to remotely measure the subject's breath rate without any significant limitations on the subject's behavior. --- ### Transmit Weights, Not Features: Orthogonal-Basis Aided Wireless Point-Cloud Transmission Authors: Junlin Chang, Yubo Han, Hang Yue, John S Thompson, Rongke Liu Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03819v2 The widespread adoption of depth sensors has substantially lowered the barrier to point-cloud acquisition. This letter proposes a semantic wireless transmission framework for three dimension (3D) point clouds built on Deep Joint Source - Channel Coding (DeepJSCC). Instead of sending raw features, the transmitter predicts combination weights over a receiver-side semantic orthogonal feature pool, enabling compact representations and robust reconstruction. A folding-based decoder deforms a 2D grid into 3D, enforcing manifold continuity while preserving geometric fidelity. Trained with Chamfer Distance (CD) and an orthogonality regularizer, the system is evaluated on ModelNet40 across varying Signal-to-Noise Ratios (SNRs) and bandwidths. Results show performance on par with SEmantic Point cloud Transmission (SEPT) at high bandwidth and clear gains in bandwidth-constrained regimes, with consistent improvements in both Peak Signal-to-Noise Ratio (PSNR) and CD. Ablation experiments confirm the benefits of orthogonalization and the folding prior. --- ### Transmit Weights, Not Features: Orthogonal-Basis Aided Wireless Point-Cloud Transmission Authors: Junlin Chang, Yubo Han, Hnag Yue, John S Thompson, Rongke Liu Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03819v1 The widespread adoption of depth sensors has substantially lowered the barrier to point-cloud acquisition. This letter proposes a semantic wireless transmission framework for three dimension (3D) point clouds built on Deep Joint Source - Channel Coding (DeepJSCC). Instead of sending raw features, the transmitter predicts combination weights over a receiver-side semantic orthogonal feature pool, enabling compact representations and robust reconstruction. A folding-based decoder deforms a 2D grid into 3D, enforcing manifold continuity while preserving geometric fidelity. Trained with Chamfer Distance (CD) and an orthogonality regularizer, the system is evaluated on ModelNet40 across varying Signal-to-Noise Ratios (SNRs) and bandwidths. Results show performance on par with SEmantic Point cloud Transmission (SEPT) at high bandwidth and clear gains in bandwidth-constrained regimes, with consistent improvements in both Peak Signal-to-Noise Ratio (PSNR) and CD. Ablation experiments confirm the benefits of orthogonalization and the folding prior. --- ### Empirical Prompt Engineering for Construct Identification with Large Language Models Authors: Kylie L. Anglin, Stephanie Milan, Brittney Hernandez, Claudia Ventura Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03818v3 Due to their architecture and vast pre-training data, large language models (LLMs) demonstrate strong performance on text classification tasks. However, LLM classifications are highly responsive to prompt wording, particularly, as we show, in domains like psychology, where constructs are often latent, complex, and theory driven. Here, we present and evaluate a systematic framework for improving psychological construct identification through prompt engineering. We combinatorially generate prompts by appending random selections of multiple variants of construct definitions, task instructions, coding guidance, and examples. Empirically selecting the highest performing of these combinations in a training dataset substantially improves alignment between LLM and human classifications. In contrast, prompting techniques such as personas, chain-of-thought reasoning, and explanations provide smaller and less consistent improvements. This finding holds across multiple models and constructs. Overall, the approach we describe offers a practical, systematic, and theory-aware method for increasing the alignment between human and LLM classifications in settings where validity is critical. --- ### Improving Alignment Between Human and Machine Codes: An Empirical Assessment of Prompt Engineering for Construct Identification in Psychology Authors: Kylie L. Anglin, Stephanie Milan, Brittney Hernandez, Claudia Ventura Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03818v1 Due to their architecture and vast pre-training data, large language models (LLMs) demonstrate strong text classification performance. However, LLM output - here, the category assigned to a text - depends heavily on the wording of the prompt. While literature on prompt engineering is expanding, few studies focus on classification tasks, and even fewer address domains like psychology, where constructs have precise, theory-driven definitions that may not be well represented in pre-training data. We present an empirical framework for optimizing LLM performance for identifying constructs in texts via prompt engineering. We experimentally evaluate five prompting strategies --codebook-guided empirical prompt selection, automatic prompt engineering, persona prompting, chain-of-thought reasoning, and explanatory prompting - with zero-shot and few-shot classification. We find that persona, chain-of-thought, and explanations do not fully address performance loss accompanying a badly worded prompt. Instead, the most influential features of a prompt are the construct definition, task framing, and, to a lesser extent, the examples provided. Across three constructs and two models, the classifications most aligned with expert judgments resulted from a few-shot prompt combining codebook-guided empirical prompt selection with automatic prompt engineering. Based on our findings, we recommend that researchers generate and evaluate as many prompt variants as feasible, whether human-crafted, automatically generated, or ideally both, and select prompts and examples based on empirical performance in a training dataset, validating the final approach in a holdout set. This procedure offers a practical, systematic, and theory-driven method for optimizing LLM prompts in settings where alignment with expert judgment is critical. --- ### HieroGlyphTranslator: Automatic Recognition and Translation of Egyptian Hieroglyphs to English Authors: Ahmed Nasser, Marwan Mohamed, Alaa Sherif, Basmala Mahmoud, Shereen Yehia, Asmaa Saad, Mariam S. El-Rahmany, Ensaf H. Mohamed Published: 2025-12-03 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.03817v1 Egyptian hieroglyphs, the ancient Egyptian writing system, are composed entirely of drawings. Translating these glyphs into English poses various challenges, including the fact that a single glyph can have multiple meanings. Deep learning translation applications are evolving rapidly, producing remarkable results that significantly impact our lives. In this research, we propose a method for the automatic recognition and translation of ancient Egyptian hieroglyphs from images to English. This study utilized two datasets for classification and translation: the Morris Franken dataset and the EgyptianTranslation dataset. Our approach is divided into three stages: segmentation (using Contour and Detectron2), mapping symbols to Gardiner codes, and translation (using the CNN model). The model achieved a BLEU score of 42.2, a significant result compared to previous research. --- ### Log Probability Tracking of LLM APIs Authors: Timothée Chauvin, Erwan Le Merrer, François Taïani, Gilles Tredan Published: 2025-12-03 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2512.03816v2 When using an LLM through an API provider, users expect the served model to remain consistent over time, a property crucial for the reliability of downstream applications and the reproducibility of research. Existing audit methods are too costly to apply at regular time intervals to the wide range of available LLM APIs. This means that model updates are left largely unmonitored in practice. In this work, we show that while LLM log probabilities (logprobs) are usually non-deterministic, they can still be used as the basis for cost-effective continuous monitoring of LLM APIs. We apply a simple statistical test based on the average value of each token logprob, requesting only a single token of output. This is enough to detect changes as small as one step of fine-tuning, making this approach more sensitive than existing methods while being 1,000x cheaper. We introduce the TinyChange benchmark as a way to measure the sensitivity of audit methods in the context of small, realistic model changes. --- ### Log Probability Tracking of LLM APIs Authors: Timothée Chauvin, Erwan Le Merrer, François Taïani, Gilles Tredan Published: 2025-12-03 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2512.03816v1 When using an LLM through an API provider, users expect the served model to remain consistent over time, a property crucial for the reliability of downstream applications and the reproducibility of research. Existing audit methods are too costly to apply at regular time intervals to the wide range of available LLM APIs. This means that model updates are left largely unmonitored in practice. In this work, we show that while LLM log probabilities (logprobs) are usually non-deterministic, they can still be used as the basis for cost-effective continuous monitoring of LLM APIs. We apply a simple statistical test based on the average value of each token logprob, requesting only a single token of output. This is enough to detect changes as small as one step of fine-tuning, making this approach more sensitive than existing methods while being 1,000x cheaper. We introduce the TinyChange benchmark as a way to measure the sensitivity of audit methods in the context of small, realistic model changes. --- ### Algorithms for Boolean Matrix Factorization using Integer Programming and Heuristics Authors: Christos Kolomvakis, Thomas Bobille, Arnaud Vandaele, Nicolas Gillis Published: 2025-12-03 Categories: cs.IR, eess.SP, math.OC, stat.ML arXiv: https://arxiv.org/abs/2512.03807v2 Boolean matrix factorization (BMF) approximates a given binary input matrix as the product of two smaller binary factors. Unlike binary matrix factorization based on standard arithmetic, BMF employs the Boolean OR and AND operations for the matrix product, which improves interpretability and reduces the approximation error. It is also used in role mining and computer vision. In this paper, we first propose algorithms for BMF that perform alternating optimization (AO) of the factor matrices, where each subproblem is solved via integer programming (IP). We then design different approaches to further enhance AO-based algorithms by selecting an optimal subset of rank-one factors from multiple runs. To address the scalability limits of IP-based methods, we introduce new greedy and local-search heuristics. We also construct a new C++ data structure for Boolean vectors and matrices that is significantly faster than existing ones and is of independent interest, allowing our heuristics to scale to large datasets. We illustrate the performance of all our proposed methods and compare them with the state of the art on various real datasets, both with and without missing data, including applications in topic modeling and imaging. --- ### Algorithms for Boolean Matrix Factorization using Integer Programming and Heuristics Authors: Christos Kolomvakis, Thomas Bobille, Arnaud Vandaele, Nicolas Gillis Published: 2025-12-03 Categories: cs.IR, eess.SP, math.OC, stat.ML arXiv: https://arxiv.org/abs/2512.03807v1 Boolean matrix factorization (BMF) approximates a given binary input matrix as the product of two smaller binary factors. Unlike binary matrix factorization based on standard arithmetic, BMF employs the Boolean OR and AND operations for the matrix product, which improves interpretability and reduces the approximation error. It is also used in role mining and computer vision. In this paper, we first propose algorithms for BMF that perform alternating optimization (AO) of the factor matrices, where each subproblem is solved via integer programming (IP). We then design different approaches to further enhance AO-based algorithms by selecting an optimal subset of rank-one factors from multiple runs. To address the scalability limits of IP-based methods, we introduce new greedy and local-search heuristics. We also construct a new C++ data structure for Boolean vectors and matrices that is significantly faster than existing ones and is of independent interest, allowing our heuristics to scale to large datasets. We illustrate the performance of all our proposed methods and compare them with the state of the art on various real datasets, both with and without missing data, including applications in topic modeling and imaging. --- ### Deep Reinforcement Learning for Dynamic Algorithm Configuration: A Case Study on Optimizing OneMax with the (1+($λ$,$λ$))-GA Authors: Tai Nguyen, Phong Le, André Biedenkapp, Carola Doerr, Nguyen Dang Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03805v2 Dynamic Algorithm Configuration (DAC) studies the efficient identification of control policies for parameterized optimization algorithms. Numerous studies leverage Reinforcement Learning (RL) to address DAC challenges; however, applying RL often requires extensive domain expertise. In this work, we conduct a comprehensive study of two deep-RL algorithms--Double Deep Q-Networks (DDQN) and Proximal Policy Optimization (PPO)--for controlling the population size of the $(1+(λ,λ))$-GA on OneMax instances. Although OneMax is structurally simple, learning effective control policies for the $(1+(λ,λ))$-GA induces a highly challenging DAC landscape, making it a controlled yet demanding benchmark. Our investigation reveals two fundamental challenges limiting DDQN and PPO: scalability degradation and learning instability, traced to under-exploration and planning horizon coverage. To address under-exploration, we introduce an adaptive reward shifting mechanism that leverages reward distribution statistics to enhance DDQN exploration. This eliminates instance-specific hyperparameter tuning and ensures consistent effectiveness across problem scales. To resolve planning horizon coverage, we demonstrate that undiscounted learning succeeds in DDQN, while PPO faces fundamental variance issues necessitating alternative designs. We further show that while hyperparameter optimization enhances PPO's stability, it consistently fails to identify effective policies. Finally, DDQN with adaptive reward shifting achieves performance comparable to theoretically derived policies with vastly improved sample efficiency, outperforming prior DAC approaches by orders of magnitude. Our findings provide insights into the fundamental obstacles faced by standard deep-RL approaches in this challenging DAC setting and highlight the key methodological ingredients required for effective learning. --- ### Deep Reinforcement Learning for Dynamic Algorithm Configuration: A Case Study on Optimizing OneMax with the (1+($λ$,$λ$))-GA Authors: Tai Nguyen, Phong Le, André Biedenkapp, Carola Doerr, Nguyen Dang Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03805v1 Dynamic Algorithm Configuration (DAC) studies the efficient identification of control policies for parameterized optimization algorithms. Numerous studies have leveraged the robustness of decision-making in Reinforcement Learning (RL) to address the optimization challenges in algorithm configuration. However, applying RL to DAC is challenging and often requires extensive domain expertise. We conduct a comprehensive study of deep-RL algorithms in DAC through a systematic analysis of controlling the population size parameter of the (1+($λ$,$λ$))-GA on OneMax instances. Our investigation of DDQN and PPO reveals two fundamental challenges that limit their effectiveness in DAC: scalability degradation and learning instability. We trace these issues to two primary causes: under-exploration and planning horizon coverage, each of which can be effectively addressed through targeted solutions. To address under-exploration, we introduce an adaptive reward shifting mechanism that leverages reward distribution statistics to enhance DDQN agent exploration, eliminating the need for instance-specific hyperparameter tuning and ensuring consistent effectiveness across different problem scales. In dealing with the planning horizon coverage problem, we demonstrate that undiscounted learning effectively resolves it in DDQN, while PPO faces fundamental variance issues that necessitate alternative algorithmic designs. We further analyze the hyperparameter dependencies of PPO, showing that while hyperparameter optimization enhances learning stability, it consistently falls short in identifying effective policies across various configurations. Finally, we demonstrate that DDQN equipped with our adaptive reward shifting strategy achieves performance comparable to theoretically derived policies with vastly improved sample efficiency, outperforming prior DAC approaches by several orders of magnitude. --- ### EfficientECG: Cross-Attention with Feature Fusion for Efficient Electrocardiogram Classification Authors: Hanhui Deng, Xinglin Li, Jie Luo, Di Wu Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03804v2 Electrocardiogram is a useful diagnostic signal that can detect cardiac abnormalities by measuring the electrical activity generated by the heart. Due to its rapid, non-invasive, and richly informative characteristics, ECG has many emerging applications. In this paper, we study novel deep learning technologies to effectively manage and analyse ECG data, with the aim of building a diagnostic model, accurately and quickly, that can substantially reduce the burden on medical workers. Unlike the existing ECG models that exhibit a high misdiagnosis rate, our deep learning approaches can automatically extract the features of ECG data through end-to-end training. Specifically, we first devise EfficientECG, an accurate and lightweight classification model for ECG analysis based on the existing EfficientNet model, which can effectively handle high-frequency long-sequence ECG data with various leading types. On top of that, we next propose a cross-attention-based feature fusion model of EfficientECG for analysing multi-lead ECG data with multiple features (e.g., gender and age). Our evaluations on representative ECG datasets validate the superiority of our model against state-of-the-art works in terms of high precision, multi-feature fusion, and lightweights. ---