# Research Papers — 2025-11 · bundle page 85 of 197

> Full-text markdown of papers #4201–4250 of 9830 published in 2025-11, newest first. AI-readable mirrors of the full 9830-paper 2025-11 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-11/page/85
[← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/84.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-11/page/86.md)

### Agentic AI Systems in Electrical Power Systems Engineering: Current State-of-the-Art and Challenges

Authors: Soham Ghosh, Gaurav Mittal
Published: 2025-11-18
Categories: eess.SY, cs.AI, cs.ET
arXiv: https://arxiv.org/abs/2511.14478v4

Agentic AI systems have recently emerged as a critical and transformative approach in artificial intelligence, offering capabilities that extend far beyond traditional AI agents and contemporary generative AI models. This rapid evolution necessitates a clear conceptual and taxonomical understanding to differentiate this new paradigm. Our paper addresses this gap by providing a comprehensive review that establishes a precise definition and taxonomy for "agentic AI," with the aim of distinguishing it from previous AI paradigms. The concepts are gradually introduced, starting with a highlight of its diverse applications across the broader field of engineering. The paper then presents four detailed, state-of-the-art use case applications specifically within electrical engineering. These case studies demonstrate practical impact, ranging from an advanced agentic framework for streamlining complex power system studies and benchmarking to a novel system developed for survival analysis of dynamic pricing strategies in battery swapping stations. Finally, to ensure robust deployment, the paper provides detailed failure mode investigations. From these findings, we derive actionable recommendations for the design and implementation of safe, reliable, and accountable agentic AI systems, offering a critical resource for researchers and practitioners.

---

### Agentic AI Systems in Electrical Power Systems Engineering: Current State-of-the-Art and Challenges

Authors: Soham Ghosh, Gaurav Mittal
Published: 2025-11-18
Categories: eess.SY, cs.AI, cs.ET
arXiv: https://arxiv.org/abs/2511.14478v1

Agentic AI systems have recently emerged as a critical and transformative approach in artificial intelligence, offering capabilities that extend far beyond traditional AI agents and contemporary generative AI models. This rapid evolution necessitates a clear conceptual and taxonomical understanding to differentiate this new paradigm. Our paper addresses this gap by providing a comprehensive review that establishes a precise definition and taxonomy for "agentic AI," with the aim of distinguishing it from previous AI paradigms. The concepts are gradually introduced, starting with a highlight of its diverse applications across the broader field of engineering. The paper then presents four detailed, state-of-the-art use case applications specifically within electrical engineering. These case studies demonstrate practical impact, ranging from an advanced agentic framework for streamlining complex power system studies and benchmarking to a novel system developed for survival analysis of dynamic pricing strategies in battery swapping stations. Finally, to ensure robust deployment, the paper provides detailed failure mode investigations. From these findings, we derive actionable recommendations for the design and implementation of safe, reliable, and accountable agentic AI systems, offering a critical resource for researchers and practitioners.

---

### 2D Gaussians Spatial Transport for Point-supervised Density Regression

Authors: Miao Shang, Xiaopeng Hong
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14477v2

This paper introduces Gaussian Spatial Transport (GST), a novel framework that leverages Gaussian splatting to facilitate transport from the probability measure in the image coordinate space to the annotation map. We propose a Gaussian splatting-based method to estimate pixel-annotation correspondence, which is then used to compute a transport plan derived from Bayesian probability. To integrate the resulting transport plan into standard network optimization in typical computer vision tasks, we derive a loss function that measures discrepancy after transport. Extensive experiments on representative computer vision tasks, including crowd counting and landmark detection, validate the effectiveness of our approach. Compared to conventional optimal transport schemes, GST eliminates iterative transport plan computation during training, significantly improving efficiency. Code is available at https://github.com/infinite0522/GST.

---

### 2D Gaussians Spatial Transport for Point-supervised Density Regression

Authors: Miao Shang, Xiaopeng Hong
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14477v1

This paper introduces Gaussian Spatial Transport (GST), a novel framework that leverages Gaussian splatting to facilitate transport from the probability measure in the image coordinate space to the annotation map. We propose a Gaussian splatting-based method to estimate pixel-annotation correspondence, which is then used to compute a transport plan derived from Bayesian probability. To integrate the resulting transport plan into standard network optimization in typical computer vision tasks, we derive a loss function that measures discrepancy after transport. Extensive experiments on representative computer vision tasks, including crowd counting and landmark detection, validate the effectiveness of our approach. Compared to conventional optimal transport schemes, GST eliminates iterative transport plan computation during training, significantly improving efficiency. Code is available at https://github.com/infinite0522/GST.

---

### Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behavior

Authors: Dalia Ali, Dora Zhao, Allison Koenecke, Orestis Papakyriakopoulos
Published: 2025-11-18
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.14476v2

Although large language models (LLMs) are increasingly trained using human feedback for safety and alignment with human values, alignment decisions often overlook human social diversity. This study examines how incorporating pluralistic values affects LLM behavior by systematically evaluating demographic variation and design parameters in the alignment pipeline. We collect alignment data from US and German participants (N = 1,095 participants, 27,375 ratings) who rated LLM responses across five dimensions: Toxicity, Emotional Awareness (EA), Sensitivity, Stereotypical Bias, and Helpfulness. We fine-tuned multiple Large Language Models and Large Reasoning Models using preferences from different social groups while varying rating scales, disagreement handling methods, and optimization techniques. The results revealed systematic demographic effects: male participants rated responses 18% less toxic than female participants; conservative and Black participants rated responses 27.9% and 44% higher on EA than liberal and White participants, respectively. Models fine-tuned on group-specific preferences exhibited distinct behaviors. Technical design choices showed strong effects: the preservation of rater disagreement achieved roughly 53% greater toxicity reduction than majority voting, and 5-point scales yielded about 22% more reduction than binary formats; and Direct Preference Optimization (DPO) consistently outperformed Group Relative Policy Optimization (GRPO) in multi-value optimization. These findings represent a preliminary step in answering a critical question: How should alignment balance expert-driven and user-driven signals to ensure both safety and fair representation?

---

### Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behavior

Authors: Dalia Ali, Dora Zhao, Allison Koenecke, Orestis Papakyriakopoulos
Published: 2025-11-18
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.14476v1

Although large language models (LLMs) are increasingly trained using human feedback for safety and alignment with human values, alignment decisions often overlook human social diversity. This study examines how incorporating pluralistic values affects LLM behavior by systematically evaluating demographic variation and design parameters in the alignment pipeline. We collected alignment data from US and German participants (N = 1,095, 27,375 ratings) who rated LLM responses across five dimensions: Toxicity, Emotional Awareness (EA), Sensitivity, Stereotypical Bias, and Helpfulness. We fine-tuned multiple Large Language Models and Large Reasoning Models using preferences from different social groups while varying rating scales, disagreement handling methods, and optimization techniques. The results revealed systematic demographic effects: male participants rated responses 18% less toxic than female participants; conservative and Black participants rated responses 27.9% and 44% more emotionally aware than liberal and White participants, respectively. Models fine-tuned on group-specific preferences exhibited distinct behaviors. Technical design choices showed strong effects: the preservation of rater disagreement achieved roughly 53% greater toxicity reduction than majority voting, and 5-point scales yielded about 22% more reduction than binary formats; and Direct Preference Optimization (DPO) consistently outperformed Group Relative Policy Optimization (GRPO) in multi-value optimization. These findings represent a preliminary step in answering a critical question: How should alignment balance expert-driven and user-driven signals to ensure both safety and fair representation?

---

### Learning Subglacial Bed Topography from Sparse Radar with Physics-Guided Residuals

Authors: Bayu Adhi Tama, Jianwu Wang, Vandana Janeja, Mostafa Cham
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14473v1

Accurate subglacial bed topography is essential for ice sheet modeling, yet radar observations are sparse and uneven. We propose a physics-guided residual learning framework that predicts bed thickness residuals over a BedMachine prior and reconstructs bed from the observed surface. A DeepLabV3+ decoder over a standard encoder (e.g.,ResNet-50) is trained with lightweight physics and data terms: multi-scale mass conservation, flow-aligned total variation, Laplacian damping, non-negativity of thickness, a ramped prior-consistency term, and a masked Huber fit to radar picks modulated by a confidence map. To measure real-world generalization, we adopt leakage-safe blockwise hold-outs (vertical/horizontal) with safety buffers and report metrics only on held-out cores. Across two Greenland sub-regions, our approach achieves strong test-core accuracy and high structural fidelity, outperforming U-Net, Attention U-Net, FPN, and a plain CNN. The residual-over-prior design, combined with physics, yields spatially coherent, physically plausible beds suitable for operational mapping under domain shift.

---

### CompEvent: Complex-valued Event-RGB Fusion for Low-light Video Enhancement and Deblurring

Authors: Mingchen Zhong, Xin Lu, Dong Li, Senyan Xu, Ruixuan Jiang, Xueyang Fu, Baocai Yin
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14469v3

Low-light video deblurring poses significant challenges in applications like nighttime surveillance and autonomous driving due to dim lighting and long exposures. While event cameras offer potential solutions with superior low-light sensitivity and high temporal resolution, existing fusion methods typically employ staged strategies, limiting their effectiveness against combined low-light and motion blur degradations. To overcome this, we propose CompEvent, a complex neural network framework enabling holistic full-process fusion of event data and RGB frames for enhanced joint restoration. CompEvent features two core components: 1) Complex Temporal Alignment GRU, which utilizes complex-valued convolutions and processes video and event streams iteratively via GRU to achieve temporal alignment and continuous fusion; and 2) Complex Space-Frequency Learning module, which performs unified complex-valued signal processing in both spatial and frequency domains, facilitating deep fusion through spatial structures and system-level characteristics. By leveraging the holistic representation capability of complex-valued neural networks, CompEvent achieves full-process spatiotemporal fusion, maximizes complementary learning between modalities, and significantly strengthens low-light video deblurring capability. Extensive experiments demonstrate that CompEvent outperforms SOTA methods in addressing this challenging task.

---

### CompEvent: Complex-valued Event-RGB Fusion for Low-light Video Enhancement and Deblurring

Authors: Mingchen Zhong, Xin Lu, Dong Li, Senyan Xu, Ruixuan Jiang, Xueyang Fu, Baocai Yin
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14469v1

Low-light video deblurring poses significant challenges in applications like nighttime surveillance and autonomous driving due to dim lighting and long exposures. While event cameras offer potential solutions with superior low-light sensitivity and high temporal resolution, existing fusion methods typically employ staged strategies, limiting their effectiveness against combined low-light and motion blur degradations. To overcome this, we propose CompEvent, a complex neural network framework enabling holistic full-process fusion of event data and RGB frames for enhanced joint restoration. CompEvent features two core components: 1) Complex Temporal Alignment GRU, which utilizes complex-valued convolutions and processes video and event streams iteratively via GRU to achieve temporal alignment and continuous fusion; and 2) Complex Space-Frequency Learning module, which performs unified complex-valued signal processing in both spatial and frequency domains, facilitating deep fusion through spatial structures and system-level characteristics. By leveraging the holistic representation capability of complex-valued neural networks, CompEvent achieves full-process spatiotemporal fusion, maximizes complementary learning between modalities, and significantly strengthens low-light video deblurring capability. Extensive experiments demonstrate that CompEvent outperforms SOTA methods in addressing this challenging task. The code is available at https://github.com/YuXie1/CompEvent.

---

### nnterp: A Standardized Interface for Mechanistic Interpretability of Transformers

Authors: Clément Dumas
Published: 2025-11-18
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.14465v2

Mechanistic interpretability research requires reliable tools for analyzing transformer internals across diverse architectures. Current approaches face a fundamental tradeoff: custom implementations like TransformerLens ensure consistent interfaces but require coding a manual adaptation for each architecture, introducing numerical mismatch with the original models, while direct HuggingFace access through NNsight preserves exact behavior but lacks standardization across models. To bridge this gap, we develop nnterp, a lightweight wrapper around NNsight that provides a unified interface for transformer analysis while preserving original HuggingFace implementations. Through automatic module renaming and comprehensive validation testing, nnterp enables researchers to write intervention code once and deploy it across 50+ model variants spanning 16 architecture families. The library includes built-in implementations of common interpretability methods (logit lens, patchscope, activation steering) and provides direct access to attention probabilities for models that support it. By packaging validation tests with the library, researchers can verify compatibility with custom models locally. nnterp bridges the gap between correctness and usability in mechanistic interpretability tooling.

---

### nnterp: A Standardized Interface for Mechanistic Interpretability of Transformers

Authors: Clément Dumas
Published: 2025-11-18
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.14465v1

Mechanistic interpretability research requires reliable tools for analyzing transformer internals across diverse architectures. Current approaches face a fundamental tradeoff: custom implementations like TransformerLens ensure consistent interfaces but require coding a manual adaptation for each architecture, introducing numerical mismatch with the original models, while direct HuggingFace access through NNsight preserves exact behavior but lacks standardization across models. To bridge this gap, we develop nnterp, a lightweight wrapper around NNsight that provides a unified interface for transformer analysis while preserving original HuggingFace implementations. Through automatic module renaming and comprehensive validation testing, nnterp enables researchers to write intervention code once and deploy it across 50+ model variants spanning 16 architecture families. The library includes built-in implementations of common interpretability methods (logit lens, patchscope, activation steering) and provides direct access to attention probabilities for models that support it. By packaging validation tests with the library, researchers can verify compatibility with custom models locally. nnterp bridges the gap between correctness and usability in mechanistic interpretability tooling.

---

### Effective Diversification of Multi-Carousel Book Recommendation

Authors: Daniël Wilten, Gideon Maillette de Buy Wenniger, Arjen Hommersom, Paul Lucassen, Emiel Poortman
Published: 2025-11-18
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2511.14461v1

Using multiple carousels, lists that wrap around and can be scrolled, is the basis for offering content in most contemporary movie streaming platforms. Carousels allow for highlighting different aspects of users' taste, that fall in categories such as genres and authors. However, while carousels offer structure and greater ease of navigation, they alone do not increase diversity in recommendations, while this is essential to keep users engaged. In this work we propose several approaches to effectively increase item diversity within the domain of book recommendations, on top of a collaborative filtering algorithm. These approaches are intended to improve book recommendations in the web catalogs of public libraries. Furthermore, we introduce metrics to evaluate the resulting strategies, and show that the proposed system finds a suitable balance between accuracy and beyond-accuracy aspects.

---

### Agent-R1: A Unified and Modular Framework for Agentic Reinforcement Learning

Authors: Mingyue Cheng, Shuo Yu, Daoyu Wang, Qingchuan Li, Xiaoyu Tao, Jie Ouyang, Yucong Luo, Yitong Zhou, Qi Liu, Enhong Chen
Published: 2025-11-18
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.14460v2

Large language models (LLMs) have rapidly evolved from single-turn text generators into the foundation of increasingly capable agents. As these agents take on more complex reasoning, decision making, tool use, and long-horizon tasks, reinforcement learning (RL) is becoming increasingly important for shaping their behavior. This shift is especially visible in agentic RL, where models must interact with tools and environments across multiple rounds rather than produce a single standalone response. In this regime, the usual view of a trajectory as one ever-growing token sequence becomes increasingly inadequate: it makes context evolution rigid and creates representation mismatches between rollout and training. This paper presents Agent-R1, a unified and modular framework for agentic RL built around step-level trajectory representation, flexible context management, and layered interfaces for workflows, environments and optimization. The key idea is to treat each interaction step as the basic reinforcement-learning transition, while keeping the optimization layer flexible: once the interaction is modeled at the step level, the framework can support token-level credit assignment, step-level credit assignment, or other compatible designs. These design choices make the framework compatible with a range of optimization strategies rather than tying it to a single algorithm. Together, these components provide a principled, extensible, and reusable substrate for agentic RL.

---

### Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning

Authors: Mingyue Cheng, Jie Ouyang, Shuo Yu, Ruiran Yan, Yucong Luo, Zirui Liu, Daoyu Wang, Qi Liu, Enhong Chen
Published: 2025-11-18
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.14460v1

Large Language Models (LLMs) are increasingly being explored for building Agents capable of active environmental interaction (e.g., via tool use) to solve complex problems. Reinforcement Learning (RL) is considered a key technology with significant potential for training such Agents; however, the effective application of RL to LLM Agents is still in its nascent stages and faces considerable challenges. Currently, this emerging field lacks in-depth exploration into RL approaches specifically tailored for the LLM Agent context, alongside a scarcity of flexible and easily extensible training frameworks designed for this purpose. To help advance this area, this paper first revisits and clarifies Reinforcement Learning methodologies for LLM Agents by systematically extending the Markov Decision Process (MDP) framework to comprehensively define the key components of an LLM Agent. Secondly, we introduce Agent-R1, a modular, flexible, and user-friendly training framework for RL-based LLM Agents, designed for straightforward adaptation across diverse task scenarios and interactive environments. We conducted experiments on Multihop QA benchmark tasks, providing initial validation for the effectiveness of our proposed methods and framework.

---

### Analyzing the Impact of Participant Failures in Cross-Silo Federated Learning

Authors: Fabian Stricker, David Bermbach, Christian Zirpins
Published: 2025-11-18
Categories: cs.DC, cs.AI
arXiv: https://arxiv.org/abs/2511.14456v1

Federated learning (FL) is a new paradigm for training machine learning (ML) models without sharing data. While applying FL in cross-silo scenarios, where organizations collaborate, it is necessary that the FL system is reliable; however, participants can fail due to various reasons (e.g., communication issues or misconfigurations). In order to provide a reliable system, it is necessary to analyze the impact of participant failures. While this problem received attention in cross-device FL where mobile devices with limited resources participate, there is comparatively little research in cross-silo FL. Therefore, we conduct an extensive study for analyzing the impact of participant failures on the model quality in the context of inter-organizational cross-silo FL with few participants. In our study, we focus on analyzing generally influential factors such as the impact of the timing and the data as well as the impact on the evaluation, which is important for deciding, if the model should be deployed. We show that under high skews the evaluation is optimistic and hides the real impact. Furthermore, we demonstrate that the timing impacts the quality of the trained model. Our results offer insights for researchers and software architects aiming to build robust FL systems.

---

### Nonparametric estimation of conditional probability distributions using a generative approach based on conditional push-forward neural networks

Authors: Nicola Rares Franco, Lorenzo Tedesco
Published: 2025-11-18
Categories: cs.LG, stat.ME
arXiv: https://arxiv.org/abs/2511.14455v3

We introduce conditional push-forward neural networks (CPFN), a generative framework for conditional distribution estimation. Instead of directly modeling the conditional density $f_{Y|X}$, CPFN learns a stochastic map $\varphi=\varphi(x,u)$ such that $\varphi(x,U)$ and $Y|X=x$ follow approximately the same law, with $U$ a suitable random vector of pre-defined latent variables. This enables efficient conditional sampling and straightforward estimation of conditional statistics through Monte Carlo methods. The model is trained via an objective function derived from a Kullback-Leibler formulation, without requiring invertibility or adversarial training. We establish a near-asymptotic consistency result and demonstrate experimentally that CPFN can achieve performance competitive with, or even superior to, state-of-the-art methods, including kernel estimators, tree-based algorithms, and popular deep learning techniques, all while remaining lightweight and easy to train.

---

### Nonparametric estimation of conditional probability distributions using a generative approach based on conditional push-forward neural networks

Authors: Nicola Rares Franco, Lorenzo Tedesco
Published: 2025-11-18
Categories: cs.LG, stat.ME
arXiv: https://arxiv.org/abs/2511.14455v1

We introduce conditional push-forward neural networks (CPFN), a generative framework for conditional distribution estimation. Instead of directly modeling the conditional density $f_{Y|X}$, CPFN learns a stochastic map $\varphi=\varphi(x,u)$ such that $\varphi(x,U)$ and $Y|X=x$ follow approximately the same law, with $U$ a suitable random vector of pre-defined latent variables. This enables efficient conditional sampling and straightforward estimation of conditional statistics through Monte Carlo methods. The model is trained via an objective function derived from a Kullback-Leibler formulation, without requiring invertibility or adversarial training. We establish a near-asymptotic consistency result and demonstrate experimentally that CPFN can achieve performance competitive with, or even superior to, state-of-the-art methods, including kernel estimators, tree-based algorithms, and popular deep learning techniques, all while remaining lightweight and easy to train.

---

### Hybrid Modeling of Photoplethysmography for Non-invasive Monitoring of Cardiovascular Parameters

Authors: Emanuele Palumbo, Sorawit Saengkyongam, Maria R. Cervera, Jens Behrmann, Andrew C. Miller, Guillermo Sapiro, Christina Heinze-Deml, Antoine Wehenkel
Published: 2025-11-18
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.14452v1

Continuous cardiovascular monitoring can play a key role in precision health. However, some fundamental cardiac biomarkers of interest, including stroke volume and cardiac output, require invasive measurements, e.g., arterial pressure waveforms (APW). As a non-invasive alternative, photoplethysmography (PPG) measurements are routinely collected in hospital settings. Unfortunately, the prediction of key cardiac biomarkers from PPG instead of APW remains an open challenge, further complicated by the scarcity of annotated PPG measurements. As a solution, we propose a hybrid approach that uses hemodynamic simulations and unlabeled clinical data to estimate cardiovascular biomarkers directly from PPG signals. Our hybrid model combines a conditional variational autoencoder trained on paired PPG-APW data with a conditional density estimator of cardiac biomarkers trained on labeled simulated APW segments. As a key result, our experiments demonstrate that the proposed approach can detect fluctuations of cardiac output and stroke volume and outperform a supervised baseline in monitoring temporal changes in these biomarkers.

---

### DIR-TIR: Dialog-Iterative Refinement for Text-to-Image Retrieval

Authors: Zongwei Zhen, Biqing Zeng
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14449v1

This paper addresses the task of interactive, conversational text-to-image retrieval. Our DIR-TIR framework progressively refines the target image search through two specialized modules: the Dialog Refiner Module and the Image Refiner Module. The Dialog Refiner actively queries users to extract essential information and generate increasingly precise descriptions of the target image. Complementarily, the Image Refiner identifies perceptual gaps between generated images and user intentions, strategically reducing the visual-semantic discrepancy. By leveraging multi-turn dialogues, DIR-TIR provides superior controllability and fault tolerance compared to conventional single-query methods, significantly improving target image hit accuracy. Comprehensive experiments across diverse image datasets demonstrate our dialogue-based approach substantially outperforms initial-description-only baselines, while the synergistic module integration achieves both higher retrieval precision and enhanced interactive experience.

---

### Agentic Video Intelligence: A Flexible Framework for Advanced Video Exploration and Understanding

Authors: Hong Gao, Yiming Bao, Xuezhen Tu, Yutong Xu, Yue Jin, Yiyang Mu, Bin Zhong, Linan Yue, Min-Ling Zhang
Published: 2025-11-18
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.14446v1

Video understanding requires not only visual recognition but also complex reasoning. While Vision-Language Models (VLMs) demonstrate impressive capabilities, they typically process videos largely in a single-pass manner with limited support for evidence revisit and iterative refinement. While recently emerging agent-based methods enable long-horizon reasoning, they either depend heavily on expensive proprietary models or require extensive agentic RL training. To overcome these limitations, we propose Agentic Video Intelligence (AVI), a flexible and training-free framework that can mirror human video comprehension through system-level design and optimization. AVI introduces three key innovations: (1) a human-inspired three-phase reasoning process (Retrieve-Perceive-Review) that ensures both sufficient global exploration and focused local analysis, (2) a structured video knowledge base organized through entity graphs, along with multi-granularity integrated tools, constituting the agent's interaction environment, and (3) an open-source model ensemble combining reasoning LLMs with lightweight base CV models and VLM, eliminating dependence on proprietary APIs or RL training. Experiments on LVBench, VideoMME-Long, LongVideoBench, and Charades-STA demonstrate that AVI achieves competitive performance while offering superior interpretability.

---

### Tell Me: An LLM-powered Mental Well-being Assistant with RAG, Synthetic Dialogue Generation, and Agentic Planning

Authors: Trishala Jayesh Ahalpara
Published: 2025-11-18
Categories: cs.CL, cs.AI, cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2511.14445v2

We present Tell Me, a mental well-being system that leverages advances in large language models to provide accessible, context-aware support for users and researchers. The system integrates three components: (i) a retrieval-augmented generation (RAG) assistant for personalized, knowledge-grounded dialogue; (ii) a synthetic client-therapist dialogue generator conditioned on client profiles to facilitate research on therapeutic language and data augmentation; and (iii) a Well-being AI crew, implemented with CrewAI, that produces weekly self-care plans and guided meditation audio. The system is designed as a reflective space for emotional processing rather than a substitute for professional therapy. It illustrates how conversational assistants can lower barriers to support, complement existing care, and broaden access to mental health resources. To address the shortage of confidential therapeutic data, we introduce synthetic client-therapist dialogue generation conditioned on client profiles. Finally, the planner demonstrates an innovative agentic workflow for dynamically adaptive, personalized self-care, bridging the limitations of static well-being tools. We describe the architecture, demonstrate its functionalities, and report evaluation of the RAG assistant in curated well-being scenarios using both automatic LLM-based judgments and a human-user study. This work highlights opportunities for interdisciplinary collaboration between NLP researchers and mental health professionals to advance responsible innovation in human-AI interaction for well-being.

---

### Tell Me: An LLM-powered Mental Well-being Assistant with RAG, Synthetic Dialogue Generation, and Agentic Planning

Authors: Trishala Jayesh Ahalpara
Published: 2025-11-18
Categories: cs.CL, cs.AI, cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2511.14445v1

We present Tell Me, a mental well-being system that leverages advances in large language models to provide accessible, context-aware support for users and researchers. The system integrates three components: (i) a retrieval-augmented generation (RAG) assistant for personalized, knowledge-grounded dialogue; (ii) a synthetic client-therapist dialogue generator conditioned on client profiles to facilitate research on therapeutic language and data augmentation; and (iii) a Well-being AI crew, implemented with CrewAI, that produces weekly self-care plans and guided meditation audio. The system is designed as a reflective space for emotional processing rather than a substitute for professional therapy. It illustrates how conversational assistants can lower barriers to support, complement existing care, and broaden access to mental health resources. To address the shortage of confidential therapeutic data, we introduce synthetic client-therapist dialogue generation conditioned on client profiles. Finally, the planner demonstrates an innovative agentic workflow for dynamically adaptive, personalized self-care, bridging the limitations of static well-being tools. We describe the architecture, demonstrate its functionalities, and report evaluation of the RAG assistant in curated well-being scenarios using both automatic LLM-based judgments and a human-user study. This work highlights opportunities for interdisciplinary collaboration between NLP researchers and mental health professionals to advance responsible innovation in human-AI interaction for well-being.

---

### Skewness-Robust Causal Discovery in Location-Scale Noise Models

Authors: Daniel Klippert, Alexander Marx
Published: 2025-11-18
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2511.14441v2

To distinguish Markov equivalent graphs in causal discovery, it is necessary to restrict the structural causal model. Crucially, we need to be able to distinguish cause $X$ from effect $Y$ in bivariate models, that is, distinguish the two graphs $X \to Y$ and $Y \to X$. Location-scale noise models (LSNMs), in which the effect $Y$ is modeled based on the cause $X$ as $Y = f(X) + g(X)N$, form a flexible class of models that is general and identifiable in most cases. Estimating these models for arbitrary noise terms $N$, however, is challenging. Therefore, practical estimators are typically restricted to symmetric distributions, such as the normal distribution. As we showcase in this paper, when $N$ is a skewed random variable, which is likely in real-world domains, the reliability of these approaches decreases. To approach this limitation, we propose SkewD, a likelihood-based algorithm for bivariate causal discovery under LSNMs with skewed noise distributions. SkewD extends the usual normal-distribution framework to the skew-normal setting, enabling reliable inference under symmetric and skewed noise. For parameter estimation, we employ a combination of a heuristic search and an expectation conditional maximization algorithm. We evaluate SkewD on novel synthetically generated datasets with skewed noise as well as established benchmark datasets. Throughout our experiments, SkewD exhibits a strong performance and, in comparison to prior work, remains robust under high skewness.

---

### Skewness-Robust Causal Discovery in Location-Scale Noise Models

Authors: Daniel Klippert, Alexander Marx
Published: 2025-11-18
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2511.14441v1

To distinguish Markov equivalent graphs in causal discovery, it is necessary to restrict the structural causal model. Crucially, we need to be able to distinguish cause $X$ from effect $Y$ in bivariate models, that is, distinguish the two graphs $X \to Y$ and $Y \to X$. Location-scale noise models (LSNMs), in which the effect $Y$ is modeled based on the cause $X$ as $Y = f(X) + g(X)N$, form a flexible class of models that is general and identifiable in most cases. Estimating these models for arbitrary noise terms $N$, however, is challenging. Therefore, practical estimators are typically restricted to symmetric distributions, such as the normal distribution. As we showcase in this paper, when $N$ is a skewed random variable, which is likely in real-world domains, the reliability of these approaches decreases. To approach this limitation, we propose SkewD, a likelihood-based algorithm for bivariate causal discovery under LSNMs with skewed noise distributions. SkewD extends the usual normal-distribution framework to the skew-normal setting, enabling reliable inference under symmetric and skewed noise. For parameter estimation, we employ a combination of a heuristic search and an expectation conditional maximization algorithm. We evaluate SkewD on novel synthetically generated datasets with skewed noise as well as established benchmark datasets. Throughout our experiments, SkewD exhibits a strong performance and, in comparison to prior work, remains robust under high skewness.

---

### Learning to See Through a Baby's Eyes: Early Visual Diets Enable Robust Visual Intelligence in Humans and Machines

Authors: Yusen Cai, Qing Lin, Bhargava Satya Nunna, Mengmi Zhang
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14440v2

Newborns perceive the world with low-acuity, color-degraded, and temporally continuous vision, which gradually sharpens as infants develop. To explore the ecological advantages of such staged "visual diets", we train self-supervised learning (SSL) models on object-centric videos under constraints that simulate infant vision: grayscale-to-color (C), blur-to-sharp (A), and preserved temporal continuity (T)-collectively termed CATDiet. For evaluation, we establish a comprehensive benchmark across ten datasets, covering clean and corrupted image recognition, texture-shape cue conflict tests, silhouette recognition, depth-order classification, and the visual cliff paradigm. All CATDiet variants demonstrate enhanced robustness in object recognition, despite being trained solely on object-centric videos. Remarkably, models also exhibit biologically aligned developmental patterns, including neural plasticity changes mirroring synaptic density in macaque V1 and behaviors resembling infants' visual cliff responses. Building on these insights, CombDiet initializes SSL with CATDiet before standard training while preserving temporal continuity. Trained on object-centric or head-mounted infant videos, CombDiet outperforms standard SSL on both in-domain and out-of-domain object recognition and depth perception. Together, these results suggest that the developmental progression of early infant visual experience offers a powerful reverse-engineering framework for understanding the emergence of robust visual intelligence in machines. All code, data, and models are available at Github.

---

### Learning to See Through a Baby's Eyes: Early Visual Diets Enable Robust Visual Intelligence in Humans and Machines

Authors: Yusen Cai, Bhargava Satya Nunna, Qing Lin, Mengmi Zhang
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14440v1

Newborns perceive the world with low-acuity, color-degraded, and temporally continuous vision, which gradually sharpens as infants develop. To explore the ecological advantages of such staged "visual diets", we train self-supervised learning (SSL) models on object-centric videos under constraints that simulate infant vision: grayscale-to-color (C), blur-to-sharp (A), and preserved temporal continuity (T)-collectively termed CATDiet. For evaluation, we establish a comprehensive benchmark across ten datasets, covering clean and corrupted image recognition, texture-shape cue conflict tests, silhouette recognition, depth-order classification, and the visual cliff paradigm. All CATDiet variants demonstrate enhanced robustness in object recognition, despite being trained solely on object-centric videos. Remarkably, models also exhibit biologically aligned developmental patterns, including neural plasticity changes mirroring synaptic density in macaque V1 and behaviors resembling infants' visual cliff responses. Building on these insights, CombDiet initializes SSL with CATDiet before standard training while preserving temporal continuity. Trained on object-centric or head-mounted infant videos, CombDiet outperforms standard SSL on both in-domain and out-of-domain object recognition and depth perception. Together, these results suggest that the developmental progression of early infant visual experience offers a powerful reverse-engineering framework for understanding the emergence of robust visual intelligence in machines. All code, data, and models will be publicly released.

---

### MedBench v4: A Robust and Scalable Benchmark for Evaluating Chinese Medical Language Models, Multimodal Models, and Intelligent Agents

Authors: Jinru Ding, Lu Lu, Chao Ding, Mouxiao Bian, Jiayuan Chen, Wenrao Pang, Ruiyao Chen, Xinwei Peng, Renjie Lu, Sijie Ren, Guanxu Zhu, Xiaoqin Wu et al.
Published: 2025-11-18
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.14439v2

Recent advances in medical large language models (LLMs), multimodal models, and agents demand evaluation frameworks that reflect real clinical workflows and safety constraints. We present MedBench v4, a nationwide, cloud-based benchmarking infrastructure comprising over 700,000 expert-curated tasks spanning 24 primary and 91 secondary specialties, with dedicated tracks for LLMs, multimodal models, and agents. Items undergo multi-stage refinement and multi-round review by clinicians from more than 500 institutions, and open-ended responses are scored by an LLM-as-a-judge calibrated to human ratings. We evaluate 15 frontier models. Base LLMs reach a mean overall score of 54.1/100 (best: Claude Sonnet 4.5, 62.5/100), but safety and ethics remain low (18.4/100). Multimodal models perform worse overall (mean 47.5/100; best: GPT-5, 54.9/100), with solid perception yet weaker cross-modal reasoning. Agents built on the same backbones substantially improve end-to-end performance (mean 79.8/100), with Claude Sonnet 4.5-based agents achieving up to 85.3/100 overall and 88.9/100 on safety tasks. MedBench v4 thus reveals persisting gaps in multimodal reasoning and safety for base models, while showing that governance-aware agentic orchestration can markedly enhance benchmarked clinical readiness without sacrificing capability. By aligning tasks with Chinese clinical guidelines and regulatory priorities, the platform offers a practical reference for hospitals, developers, and policymakers auditing medical AI.

---

### MedBench v4: A Robust and Scalable Benchmark for Evaluating Chinese Medical Language Models, Multimodal Models, and Intelligent Agents

Authors: Jinru Ding, Lu Lu, Chao Ding, Mouxiao Bian, Jiayuan Chen, Renjie Lu, Wenrao Pang, Xiaoqin Wu, Zhiqiang Liu, Luyi Jiang, Bing Han, Yunqiu Wang et al.
Published: 2025-11-18
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.14439v1

Recent advances in medical large language models (LLMs), multimodal models, and agents demand evaluation frameworks that reflect real clinical workflows and safety constraints. We present MedBench v4, a nationwide, cloud-based benchmarking infrastructure comprising over 700,000 expert-curated tasks spanning 24 primary and 91 secondary specialties, with dedicated tracks for LLMs, multimodal models, and agents. Items undergo multi-stage refinement and multi-round review by clinicians from more than 500 institutions, and open-ended responses are scored by an LLM-as-a-judge calibrated to human ratings. We evaluate 15 frontier models. Base LLMs reach a mean overall score of 54.1/100 (best: Claude Sonnet 4.5, 62.5/100), but safety and ethics remain low (18.4/100). Multimodal models perform worse overall (mean 47.5/100; best: GPT-5, 54.9/100), with solid perception yet weaker cross-modal reasoning. Agents built on the same backbones substantially improve end-to-end performance (mean 79.8/100), with Claude Sonnet 4.5-based agents achieving up to 85.3/100 overall and 88.9/100 on safety tasks. MedBench v4 thus reveals persisting gaps in multimodal reasoning and safety for base models, while showing that governance-aware agentic orchestration can markedly enhance benchmarked clinical readiness without sacrificing capability. By aligning tasks with Chinese clinical guidelines and regulatory priorities, the platform offers a practical reference for hospitals, developers, and policymakers auditing medical AI.

---

### Watchdogs and Oracles: Runtime Verification Meets Large Language Models for Autonomous Systems

Authors: Angelo Ferrando
Published: 2025-11-18
Categories: cs.SE, cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2511.14435v1

Assuring the safety and trustworthiness of autonomous systems is particularly difficult when learning-enabled components and open environments are involved. Formal methods provide strong guarantees but depend on complete models and static assumptions. Runtime verification (RV) complements them by monitoring executions at run time and, in its predictive variants, by anticipating potential violations. Large language models (LLMs), meanwhile, excel at translating natural language into formal artefacts and recognising patterns in data, yet they remain error-prone and lack formal guarantees. This vision paper argues for a symbiotic integration of RV and LLMs. RV can serve as a guardrail for LLM-driven autonomy, while LLMs can extend RV by assisting specification capture, supporting anticipatory reasoning, and helping to handle uncertainty. We outline how this mutual reinforcement differs from existing surveys and roadmaps, discuss challenges and certification implications, and identify future research directions towards dependable autonomy.

---

### Context-aware, Ante-hoc Explanations of Driving Behaviour

Authors: Dominik Grundt, Ishan Saxena, Malte Petersen, Bernd Westphal, Eike Möhlmann
Published: 2025-11-18
Categories: cs.LO, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2511.14428v1

Autonomous vehicles (AVs) must be both safe and trustworthy to gain social acceptance and become a viable option for everyday public transportation. Explanations about the system behaviour can increase safety and trust in AVs. Unfortunately, explaining the system behaviour of AI-based driving functions is particularly challenging, as decision-making processes are often opaque. The field of Explainability Engineering tackles this challenge by developing explanation models at design time. These models are designed from system design artefacts and stakeholder needs to develop correct and good explanations. To support this field, we propose an approach that enables context-aware, ante-hoc explanations of (un)expectable driving manoeuvres at runtime. The visual yet formal language Traffic Sequence Charts is used to formalise explanation contexts, as well as corresponding (un)expectable driving manoeuvres. A dedicated runtime monitoring enables context-recognition and ante-hoc presentation of explanations at runtime. In combination, we aim to support the bridging of correct and good explanations. Our method is demonstrated in a simulated overtaking.

---

### Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

Authors: Rickmer Krohn, Vignesh Prasad, Gabriele Tiboni, Georgia Chalvatzaki
Published: 2025-11-18
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2511.14427v4

Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception. However, Reinforcement Learning agents struggle to learn in such multisensory settings, especially amidst sensory noise and dynamic changes. We propose MultiSensory Dynamic Pretraining (MSDP), a novel framework for learning expressive multisensory representations tailored for task-oriented policy learning. MSDP is based on masked autoencoding and trains a transformer-based encoder by reconstructing multisensory observations from only a subset of sensor embeddings, leading to cross-modal prediction and sensor fusion. For downstream policy learning, we introduce a novel asymmetric architecture, where a cross-attention mechanism allows the critic to extract dynamic, task-specific features from the frozen embeddings, while the actor receives a stable pooled representation to guide its actions. Our method demonstrates accelerated learning and robust performance under diverse perturbations, including sensor noise, and changes in object dynamics. Evaluations in multiple challenging, contact-rich robot manipulation tasks in simulation and the real world showcase the effectiveness of MSDP. Our approach exhibits strong robustness to perturbations and achieves high success rates on the real robot with as few as 6,000 online interactions, offering a simple yet powerful solution for complex multisensory robotic control. Website: https://msdp-pearl.github.io/

---

### Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

Authors: Rickmer Krohn, Vignesh Prasad, Gabriele Tiboni, Georgia Chalvatzaki
Published: 2025-11-18
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2511.14427v1

Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception. However, Reinforcement Learning agents struggle to learn in such multisensory settings, especially amidst sensory noise and dynamic changes. We propose MultiSensory Dynamic Pretraining (MSDP), a novel framework for learning expressive multisensory representations tailored for task-oriented policy learning. MSDP is based on masked autoencoding and trains a transformer-based encoder by reconstructing multisensory observations from only a subset of sensor embeddings, leading to cross-modal prediction and sensor fusion. For downstream policy learning, we introduce a novel asymmetric architecture, where a cross-attention mechanism allows the critic to extract dynamic, task-specific features from the frozen embeddings, while the actor receives a stable pooled representation to guide its actions. Our method demonstrates accelerated learning and robust performance under diverse perturbations, including sensor noise, and changes in object dynamics. Evaluations in multiple challenging, contact-rich robot manipulation tasks in simulation and the real world showcase the effectiveness of MSDP. Our approach exhibits strong robustness to perturbations and achieves high success rates on the real robot with as few as 6,000 online interactions, offering a simple yet powerful solution for complex multisensory robotic control.

---

### MiAD: Mirage Atom Diffusion for De Novo Crystal Generation

Authors: Andrey Okhotin, Maksim Nakhodnov, Nikita Kazeev, Mikhail Lazarev, Andrey E Ustyuzhanin, Dmitry Vetrov
Published: 2025-11-18
Categories: cs.LG, cond-mat.mtrl-sci, cs.AI, physics.comp-ph
arXiv: https://arxiv.org/abs/2511.14426v2

In recent years, diffusion-based models have demonstrated exceptional performance in searching for simultaneously stable, unique, and novel (S.U.N.) crystalline materials. However, most of these models don't have the ability to change the number of atoms in the crystal during the generation process, which limits the variability of model sampling trajectories. In this paper, we demonstrate the severity of this restriction and introduce a simple yet powerful technique, mirage infusion, which enables diffusion models to change the state of the atoms that make up the crystal from existent to non-existent (mirage) and vice versa. We show that this technique improves model quality by up to x2.5 compared to the same model without this modification. The resulting model, Mirage Atom Diffusion (MiAD), is an equivariant joint diffusion model for de novo crystal generation that is capable of altering the number of atoms during the generation process. MiAD achieves an 8.2% S.U.N. rate on the MP-20 dataset, which substantially exceeds existing state-of-the-art approaches. Code: https://github.com/andrey-okhotin/miad.git

---

### MiAD: Mirage Atom Diffusion for De Novo Crystal Generation

Authors: Andrey Okhotin, Maksim Nakhodnov, Nikita Kazeev, Andrey E Ustyuzhanin, Dmitry Vetrov
Published: 2025-11-18
Categories: cs.LG, cond-mat.mtrl-sci, cs.AI, physics.comp-ph
arXiv: https://arxiv.org/abs/2511.14426v1

In recent years, diffusion-based models have demonstrated exceptional performance in searching for simultaneously stable, unique, and novel (S.U.N.) crystalline materials. However, most of these models don't have the ability to change the number of atoms in the crystal during the generation process, which limits the variability of model sampling trajectories. In this paper, we demonstrate the severity of this restriction and introduce a simple yet powerful technique, mirage infusion, which enables diffusion models to change the state of the atoms that make up the crystal from existent to non-existent (mirage) and vice versa. We show that this technique improves model quality by up to $\times2.5$ compared to the same model without this modification. The resulting model, Mirage Atom Diffusion (MiAD), is an equivariant joint diffusion model for de novo crystal generation that is capable of altering the number of atoms during the generation process. MiAD achieves an $8.2\%$ S.U.N. rate on the MP-20 dataset, which substantially exceeds existing state-of-the-art approaches. The source code can be found at \href{https://github.com/andrey-okhotin/miad.git}{\texttt{github.com/andrey-okhotin/miad}}.

---

### Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Education

Authors: Xin Yi, Yue Li, Dongsheng Shi, Linlin Wang, Xiaoling Wang, Liang He
Published: 2025-11-18
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.14423v1

Large Language Models (LLMs) are increasingly integrated into educational applications. However, they remain vulnerable to jailbreak and fine-tuning attacks, which can compromise safety alignment and lead to harmful outputs. Existing studies mainly focus on general safety evaluations, with limited attention to the unique safety requirements of educational scenarios. To address this gap, we construct EduHarm, a benchmark containing safe-unsafe instruction pairs across five representative educational scenarios, enabling systematic safety evaluation of educational LLMs. Furthermore, we propose a three-stage shield framework (TSSF) for educational LLMs that simultaneously mitigates both jailbreak and fine-tuning attacks. First, safety-aware attention realignment redirects attention toward critical unsafe tokens, thereby restoring the harmfulness feature that discriminates between unsafe and safe inputs. Second, layer-wise safety judgment identifies harmfulness features by aggregating safety cues across multiple layers to detect unsafe instructions. Finally, defense-driven dual routing separates safe and unsafe queries, ensuring normal processing for benign inputs and guarded responses for harmful ones. Extensive experiments across eight jailbreak attack strategies demonstrate that TSSF effectively strengthens safety while preventing over-refusal of benign queries. Evaluations on three fine-tuning attack datasets further show that it consistently achieves robust defense against harmful queries while maintaining preserving utility gains from benign fine-tuning.

---

### Server-Enforced Watermarking in U-Shaped Split Federated Learning

Authors: Zhengchunmin Dai, Jiaxiong Tang, Peng Sun, Honglong Chen, Liantao Wu
Published: 2025-11-18
Categories: cs.CR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.14422v2

U-shaped split federated learning (U-SFL) enables resource-constrained Internet of Things devices to collaboratively train models with an edge server while retaining raw data and labels locally. We consider model-service deployments in which a provider supplies proprietary models whose client-side segments reside on participating devices, creating risks of unauthorized copying and redistribution. Protecting these segments is challenging because the server cannot directly access client data, labels, or model parameters, and potentially malicious clients may refuse to perform watermark embedding. To address this challenge, we propose Sigil, a server-enforced watermarking framework for U-SFL. Sigil defines a secret watermark constraint in the server-visible activation space and embeds the watermark into client-side models by injecting a watermark gradient into the gradients returned during training. This mechanism requires neither access to clients' raw data and labels nor client-side watermarking operations. To limit interference with the main task and reduce detectability by gradient anomaly detectors, Sigil adaptively clips the watermark gradient relative to the main-task gradient. Experiments on two datasets and four model architectures demonstrate high watermark detection rates with limited impact on task accuracy, robustness against the evaluated removal attacks, and stealthiness against the evaluated gradient anomaly detector.

---

### Sigil: Server-Enforced Watermarking in U-Shaped Split Federated Learning via Gradient Injection

Authors: Zhengchunmin Dai, Jiaxiong Tang, Peng Sun, Honglong Chen, Liantao Wu
Published: 2025-11-18
Categories: cs.CR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.14422v1

In decentralized machine learning paradigms such as Split Federated Learning (SFL) and its variant U-shaped SFL, the server's capabilities are severely restricted. Although this enhances client-side privacy, it also leaves the server highly vulnerable to model theft by malicious clients. Ensuring intellectual property protection for such capability-limited servers presents a dual challenge: watermarking schemes that depend on client cooperation are unreliable in adversarial settings, whereas traditional server-side watermarking schemes are technically infeasible because the server lacks access to critical elements such as model parameters or labels. To address this challenge, this paper proposes Sigil, a mandatory watermarking framework designed specifically for capability-limited servers. Sigil defines the watermark as a statistical constraint on the server-visible activation space and embeds the watermark into the client model via gradient injection, without requiring any knowledge of the data. Besides, we design an adaptive gradient clipping mechanism to ensure that our watermarking process remains both mandatory and stealthy, effectively countering existing gradient anomaly detection methods and a specifically designed adaptive subspace removal attack. Extensive experiments on multiple datasets and models demonstrate Sigil's fidelity, robustness, and stealthiness.

---

### FlowRoI A Fast Optical Flow Driven Region of Interest Extraction Framework for High-Throughput Image Compression in Immune Cell Migration Analysis

Authors: Xiaowei Xu, Justin Sonneck, Hongxiao Wang, Roman Burkard, Hendrik Wohrle, Anton Grabmasier, Matthias Gunzer, Jianxu Chen
Published: 2025-11-18
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.14419v1

Autonomous migration is essential for the function of immune cells such as neutrophils and plays a pivotal role in diverse diseases. Recently, we introduced ComplexEye, a multi-lens array microscope comprising 16 independent aberration-corrected glass lenses arranged at the pitch of a 96-well plate, capable of capturing high-resolution movies of migrating cells. This architecture enables high-throughput live-cell video microscopy for migration analysis, supporting routine quantification of autonomous motility with strong potential for clinical translation. However, ComplexEye and similar high-throughput imaging platforms generate data at an exponential rate, imposing substantial burdens on storage and transmission. To address this challenge, we present FlowRoI, a fast optical-flow-based region of interest (RoI) extraction framework designed for high-throughput image compression in immune cell migration studies. FlowRoI estimates optical flow between consecutive frames and derives RoI masks that reliably cover nearly all migrating cells. The raw image and its corresponding RoI mask are then jointly encoded using JPEG2000 to enable RoI-aware compression. FlowRoI operates with high computational efficiency, achieving runtimes comparable to standard JPEG2000 and reaching an average throughput of about 30 frames per second on a modern laptop equipped with an Intel i7-1255U CPU. In terms of image quality, FlowRoI yields higher peak signal-to-noise ratio (PSNR) in cellular regions and achieves 2.0-2.2x higher compression rates at matched PSNR compared to standard JPEG2000.

---

### Toward Robust and Harmonious Adaptation for Cross-modal Retrieval

Authors: Haobin Li, Mouxing Yang, Xi Peng
Published: 2025-11-18
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.14416v1

Recently, the general-to-customized paradigm has emerged as the dominant approach for Cross-Modal Retrieval (CMR), which reconciles the distribution shift problem between the source domain and the target domain. However, existing general-to-customized CMR methods typically assume that the entire target-domain data is available, which is easily violated in real-world scenarios and thus inevitably suffer from the query shift (QS) problem. Specifically, query shift embraces the following two characteristics and thus poses new challenges to CMR. i) Online Shift: real-world queries always arrive in an online manner, rendering it impractical to access the entire query set beforehand for customization approaches; ii) Diverse Shift: even with domain customization, the CMR models struggle to satisfy queries from diverse users or scenarios, leaving an urgent need to accommodate diverse queries. In this paper, we observe that QS would not only undermine the well-structured common space inherited from the source model, but also steer the model toward forgetting the indispensable general knowledge for CMR. Inspired by the observations, we propose a novel method for achieving online and harmonious adaptation against QS, dubbed Robust adaptation with quEry ShifT (REST). To deal with online shift, REST first refines the retrieval results to formulate the query predictions and accordingly designs a QS-robust objective function on these predictions to preserve the well-established common space in an online manner. As for tackling the more challenging diverse shift, REST employs a gradient decoupling module to dexterously manipulate the gradients during the adaptation process, thus preventing the CMR model from forgetting the general knowledge. Extensive experiments on 20 benchmarks across three CMR tasks verify the effectiveness of our method against QS.

---

### Cranio-ID: Graph-Based Craniofacial Identification via Automatic Landmark Annotation in 2D Multi-View X-rays

Authors: Ravi Shankar Prasad, Nandani Sharma, Dinesh Singh
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14411v1

In forensic craniofacial identification and in many biomedical applications, craniometric landmarks are important. Traditional methods for locating landmarks are time-consuming and require specialized knowledge and expertise. Current methods utilize superimposition and deep learning-based methods that employ automatic annotation of landmarks. However, these methods are not reliable due to insufficient large-scale validation studies. In this paper, we proposed a novel framework Cranio-ID: First, an automatic annotation of landmarks on 2D skulls (which are X-ray scans of faces) with their respective optical images using our trained YOLO-pose models. Second, cross-modal matching by formulating these landmarks into graph representations and then finding semantic correspondence between graphs of these two modalities using cross-attention and optimal transport framework. Our proposed framework is validated on the S2F and CUHK datasets (CUHK dataset resembles with S2F dataset). Extensive experiments have been conducted to evaluate the performance of our proposed framework, which demonstrates significant improvements in both reliability and accuracy, as well as its effectiveness in cross-domain skull-to-face and sketch-to-face matching in forensic science.

---

### Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation

Authors: Bastien Vuillod, Pierre-Alain Moellic, Jean-Max Dutertre
Published: 2025-11-18
Categories: cs.LG, cs.CR
arXiv: https://arxiv.org/abs/2511.14406v2

Large models adaptation through Federated Learning (FL) addresses a wide range of use cases and is enabled by Parameter-Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA). However, this distributed learning paradigm faces several security threats, particularly to its integrity, such as backdoor attacks that aim to inject malicious behavior during the local training steps of certain clients. We present the first analysis of the influence of LoRA on state-of-the-art backdoor attacks targeting model adaptation in FL. Specifically, we focus on backdoor lifespan, a critical characteristic in FL, that can vary depending on the attack scenario and the attacker's ability to effectively inject the backdoor. A key finding in our experiments is that for an optimally injected backdoor, the backdoor persistence after the attack is longer when the LoRA's rank is lower. Importantly, our work highlights evaluation issues of backdoor attacks against FL and contributes to the development of more robust and fair evaluations of backdoor attacks, enhancing the reliability of risk assessments for critical FL systems. Our code is publicly available.

---

### Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation

Authors: Bastien Vuillod, Pierre-Alain Moellic, Jean-Max Dutertre
Published: 2025-11-18
Categories: cs.LG, cs.CR
arXiv: https://arxiv.org/abs/2511.14406v1

Large models adaptation through Federated Learning (FL) addresses a wide range of use cases and is enabled by Parameter-Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA). However, this distributed learning paradigm faces several security threats, particularly to its integrity, such as backdoor attacks that aim to inject malicious behavior during the local training steps of certain clients. We present the first analysis of the influence of LoRA on state-of-the-art backdoor attacks targeting model adaptation in FL. Specifically, we focus on backdoor lifespan, a critical characteristic in FL, that can vary depending on the attack scenario and the attacker's ability to effectively inject the backdoor. A key finding in our experiments is that for an optimally injected backdoor, the backdoor persistence after the attack is longer when the LoRA's rank is lower. Importantly, our work highlights evaluation issues of backdoor attacks against FL and contributes to the development of more robust and fair evaluations of backdoor attacks, enhancing the reliability of risk assessments for critical FL systems. Our code is publicly available.

---

### Language as an Anchor: Preserving Relative Visual Geometry for Domain Incremental Learning

Authors: Shuyi Geng, Tao Zhou, Yi Zhou
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14401v1

A key challenge in Domain Incremental Learning (DIL) is to continually learn under shifting distributions while preserving knowledge from previous domains. Existing methods face a fundamental dilemma. On one hand, projecting all domains into a single unified visual space leads to inter-domain interference and semantic distortion, as large shifts may vary with not only visual appearance but also underlying semantics. On the other hand, isolating domain-specific parameters causes knowledge fragmentation, creating "knowledge islands" that hamper knowledge reuse and exacerbate forgetting. To address this issue, we propose LAVA (Language-Anchored Visual Alignment), a novel DIL framework that replaces direct feature alignment with relative alignment driven by a text-based reference anchor. LAVA guides the visual representations of each incoming domain to preserve a consistent relative geometry, which is defined by mirroring the pairwise semantic similarities between the class names. This anchored geometric structure acts as a bridge across domains, enabling the retrieval of class-aware prior knowledge and facilitating robust feature aggregation. Extensive experiments on standard DIL benchmarks demonstrate that LAVA achieves significant performance improvements over state-of-the-arts. Code is available at https://github.com/ShuyiGeng/LAVA.

---

### Stage Aware Diagnosis of Diabetic Retinopathy via Ordinal Regression

Authors: Saksham Kumar, D Sridhar Aditya, T Likhil Kumar, Thulasi Bikku, Srinivasarao Thota, Chandan Kumar
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14398v1

Diabetic Retinopathy (DR) has emerged as a major cause of preventable blindness in recent times. With timely screening and intervention, the condition can be prevented from causing irreversible damage. The work introduces a state-of-the-art Ordinal Regression-based DR Detection framework that uses the APTOS-2019 fundus image dataset. A widely accepted combination of preprocessing methods: Green Channel (GC) Extraction, Noise Masking, and CLAHE, was used to isolate the most relevant features for DR classification. Model performance was evaluated using the Quadratic Weighted Kappa, with a focus on agreement between results and clinical grading. Our Ordinal Regression approach attained a QWK score of 0.8992, setting a new benchmark on the APTOS dataset.

---

### Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning

Authors: Xiuxiu Qi, Yu Yang, Jiannong Cao, Luyao Bai, Chongshan Fan, Chengtai Cao, Hongpeng Wang
Published: 2025-11-18
Categories: cs.RO, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2511.14396v5

Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.

---

### Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning

Authors: Xiuxiu Qi, Yu Yang, Jiannong Cao, Luyao Bai, Chongshan Fan, Chengtai Cao, Hongpeng Wang
Published: 2025-11-18
Categories: cs.RO, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2511.14396v1

Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.

---

### BEDLAM2.0: Synthetic Humans and Cameras in Motion

Authors: Joachim Tesch, Giorgio Becherini, Prerana Achar, Anastasios Yiannakidis, Muhammed Kocabas, Priyanka Patel, Michael J. Black
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14394v1

Inferring 3D human motion from video remains a challenging problem with many applications. While traditional methods estimate the human in image coordinates, many applications require human motion to be estimated in world coordinates. This is particularly challenging when there is both human and camera motion. Progress on this topic has been limited by the lack of rich video data with ground truth human and camera movement. We address this with BEDLAM2.0, a new dataset that goes beyond the popular BEDLAM dataset in important ways. In addition to introducing more diverse and realistic cameras and camera motions, BEDLAM2.0 increases diversity and realism of body shape, motions, clothing, hair, and 3D environments. Additionally, it adds shoes, which were missing in BEDLAM. BEDLAM has become a key resource for training 3D human pose and motion regressors today and we show that BEDLAM2.0 is significantly better, particularly for training methods that estimate humans in world coordinates. We compare state-of-the art methods trained on BEDLAM and BEDLAM2.0, and find that BEDLAM2.0 significantly improves accuracy over BEDLAM. For research purposes, we provide the rendered videos, ground truth body parameters, and camera motions. We also provide the 3D assets to which we have rights and links to those from third parties.

---

### Enhancing LLM-based Autonomous Driving with Modular Traffic Light and Sign Recognition

Authors: Fabian Schmidt, Noushiq Mohammed Kayilan Abdul Nazar, Markus Enzweiler, Abhinav Valada
Published: 2025-11-18
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.14391v1

Large Language Models (LLMs) are increasingly used for decision-making and planning in autonomous driving, showing promising reasoning capabilities and potential to generalize across diverse traffic situations. However, current LLM-based driving agents lack explicit mechanisms to enforce traffic rules and often struggle to reliably detect small, safety-critical objects such as traffic lights and signs. To address this limitation, we introduce TLS-Assist, a modular redundancy layer that augments LLM-based autonomous driving agents with explicit traffic light and sign recognition. TLS-Assist converts detections into structured natural language messages that are injected into the LLM input, enforcing explicit attention to safety-critical cues. The framework is plug-and-play, model-agnostic, and supports both single-view and multi-view camera setups. We evaluate TLS-Assist in a closed-loop setup on the LangAuto benchmark in CARLA. The results demonstrate relative driving performance improvements of up to 14% over LMDrive and 7% over BEVDriver, while consistently reducing traffic light and sign infractions. We publicly release the code and models on https://github.com/iis-esslingen/TLS-Assist.

---

### Cheating Stereo Matching in Full-scale: Physical Adversarial Attack against Binocular Depth Estimation in Autonomous Driving

Authors: Kangqiao Zhao, Shuo Huai, Xurui Song, Jun Luo
Published: 2025-11-18
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.14386v4

Though deep neural models adopted to realize the perception of autonomous driving have proven vulnerable to adversarial examples, known attacks often leverage 2D patches and target mostly monocular perception. Therefore, the effectiveness of Physical Adversarial Examples (PAEs) on stereo-based binocular depth estimation remains largely unexplored. To this end, we propose the first texture-enabled physical adversarial attack against stereo matching models in the context of autonomous driving. Our method employs a 3D PAE with global camouflage texture rather than a local 2D patch-based one, ensuring both visual consistency and attack effectiveness across different viewpoints of stereo cameras. To cope with the disparity effect of these cameras, we also propose a new 3D stereo matching rendering module that allows the PAE to be aligned with real-world positions and headings in binocular vision. We further propose a novel merging attack that seamlessly blends the target into the environment through fine-grained PAE optimization. It has significantly enhanced stealth and lethality upon existing hiding attacks that fail to get seamlessly merged into the background. Extensive evaluations show that our PAEs can successfully fool the stereo models into producing erroneous depth information.

---

### Cheating Stereo Matching in Full-scale: Physical Adversarial Attack against Binocular Depth Estimation in Autonomous Driving

Authors: Kangqiao Zhao, Shuo Huai, Xurui Song, Jun Luo
Published: 2025-11-18
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.14386v1

Though deep neural models adopted to realize the perception of autonomous driving have proven vulnerable to adversarial examples, known attacks often leverage 2D patches and target mostly monocular perception. Therefore, the effectiveness of Physical Adversarial Examples (PAEs) on stereo-based binocular depth estimation remains largely unexplored. To this end, we propose the first texture-enabled physical adversarial attack against stereo matching models in the context of autonomous driving. Our method employs a 3D PAE with global camouflage texture rather than a local 2D patch-based one, ensuring both visual consistency and attack effectiveness across different viewpoints of stereo cameras. To cope with the disparity effect of these cameras, we also propose a new 3D stereo matching rendering module that allows the PAE to be aligned with real-world positions and headings in binocular vision. We further propose a novel merging attack that seamlessly blends the target into the environment through fine-grained PAE optimization. It has significantly enhanced stealth and lethality upon existing hiding attacks that fail to get seamlessly merged into the background. Extensive evaluations show that our PAEs can successfully fool the stereo models into producing erroneous depth information.

---