# Research Papers — 2025-11 · bundle page 110 of 197 > Full-text markdown of papers #5451–5500 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/110 [← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/109.md) [Next bundle →](https://aipod.dev/papers/archive/2025-11/page/111.md) ### Collaborative Representation Learning for Alignment of Tactile, Language, and Vision Modalities Authors: Yiyun Zhou, Mingjing Xu, Jingwei Shi, Quanjiang Li, Jingyuan Chen Published: 2025-11-14 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2511.11512v1 Tactile sensing offers rich and complementary information to vision and language, enabling robots to perceive fine-grained object properties. However, existing tactile sensors lack standardization, leading to redundant features that hinder cross-sensor generalization. Moreover, existing methods fail to fully integrate the intermediate communication among tactile, language, and vision modalities. To address this, we propose TLV-CoRe, a CLIP-based Tactile-Language-Vision Collaborative Representation learning method. TLV-CoRe introduces a Sensor-Aware Modulator to unify tactile features across different sensors and employs tactile-irrelevant decoupled learning to disentangle irrelevant tactile features. Additionally, a Unified Bridging Adapter is introduced to enhance tri-modal interaction within the shared representation space. To fairly evaluate the effectiveness of tactile models, we further propose the RSS evaluation framework, focusing on Robustness, Synergy, and Stability across different methods. Experimental results demonstrate that TLV-CoRe significantly improves sensor-agnostic representation learning and cross-modal alignment, offering a new direction for multimodal tactile representation. --- ### OpenUS: A Fully Open-Source Foundation Model for Ultrasound Image Analysis via Self-Adaptive Masked Contrastive Learning Authors: Xiaoyu Zheng, Xu Chen, Awais Rauf, Qifan Fu, Benedetta Monosi, Felice Rivellese, Myles J. Lewis, Shaogang Gong, Gregory Slabaugh Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11510v1 Ultrasound (US) is one of the most widely used medical imaging modalities, thanks to its low cost, portability, real-time feedback, and absence of ionizing radiation. However, US image interpretation remains highly operator-dependent and varies significantly across anatomical regions, acquisition protocols, and device types. These variations, along with unique challenges such as speckle, low contrast, and limited standardized annotations, hinder the development of generalizable, label-efficient ultrasound AI models. In this paper, we propose OpenUS, the first reproducible, open-source ultrasound foundation model built on a large collection of public data. OpenUS employs a vision Mamba backbone, capturing both local and global long-range dependencies across the image. To extract rich features during pre-training, we introduce a novel self-adaptive masking framework that combines contrastive learning with masked image modeling. This strategy integrates the teacher's attention map with student reconstruction loss, adaptively refining clinically-relevant masking to enhance pre-training effectiveness. OpenUS also applies a dynamic learning schedule to progressively adjust the difficulty of the pre-training process. To develop the foundation model, we compile the largest to-date public ultrasound dataset comprising over 308K images from 42 publicly available datasets, covering diverse anatomical regions, institutions, imaging devices, and disease types. Our pre-trained OpenUS model can be easily adapted to specific downstream tasks by serving as a backbone for label-efficient fine-tuning. Code is available at https://github.com/XZheng0427/OpenUS. --- ### FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts Models Authors: Yonatan Dukler, Guihong Li, Deval Shah, Jiang Liu, Vikram Appia, Emad Barsoum Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11505v3 Blocking communication presents a major hurdle in running MoEs efficiently in distributed settings. To address this, we present FarSkip-Collective which modifies the architecture of modern models to enable overlapping of their computation with communication. Our approach modifies the architecture to skip connections in the model and it is unclear a priori whether the modified model architecture can remain as capable, especially for large state-of-the-art models and while modifying all of the model layers. We answer this question in the affirmative and fully convert a series of state-of-the-art models varying from 16B to 109B parameters to enable overlapping of their communication while achieving accuracy that is comparable with their original open-source releases. For example, we convert Llama 4 Scout (109B) via self-distillation and achieve average accuracy within 1% of its instruction tuned release averaged over a wide range of downstream evaluations. In addition to demonstrating retained accuracy of the large modified models, we realize the benefits of FarSkip-Collective through optimized implementations that explicitly overlap communication with computation, accelerating both training and inference in existing frameworks. For inference, we demonstrate 32.6% speedup in Time To First Token when serving a converted DeepSeek-V3 architecture with expert parallelism in SGLang and achieve 97.3% communication-computation overlap during the prefill stage. During training, our approach enables 88.9% communication overlap of the all-to-all communication collectives when pre-training DeepSeek-V3 MoE layers with expert parallelism. --- ### FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts Models Authors: Yonatan Dukler, Guihong Li, Deval Shah, Vikram Appia, Emad Barsoum Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11505v1 Blocking communication presents a major hurdle in running MoEs efficiently in distributed settings. To address this, we present FarSkip-Collective which modifies the architecture of modern models to enable overlapping of their computation with communication. Our approach modifies the architecture to skip connections in the model and it is unclear a priori whether the modified model architecture can remain as capable, especially for large state-of-the-art models and while modifying all of the model layers. We answer this question in the affirmative and fully convert a series of state-of-the-art models varying from 16B to 109B parameters to enable overlapping of their communication while achieving accuracy on par with their original open-source releases. For example, we convert Llama 4 Scout (109B) via self-distillation and achieve average accuracy within 1% of its instruction tuned release averaged across a wide range of downstream evaluations. In addition to demonstrating retained accuracy of the large modified models, we realize the benefits of FarSkip-Collective through optimized implementations that explicitly overlap communication with computation, accelerating both training and inference in existing frameworks. --- ### PAS : Prelim Attention Score for Detecting Object Hallucinations in Large Vision--Language Models Authors: Nhat Hoang-Xuan, Minh Vu, My T. Thai, Manish Bhattarai Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11502v1 Large vision-language models (LVLMs) are powerful, yet they remain unreliable due to object hallucinations. In this work, we show that in many hallucinatory predictions the LVLM effectively ignores the image and instead relies on previously generated output (prelim) tokens to infer new objects. We quantify this behavior via the mutual information between the image and the predicted object conditioned on the prelim, demonstrating that weak image dependence strongly correlates with hallucination. Building on this finding, we introduce the Prelim Attention Score (PAS), a lightweight, training-free signal computed from attention weights over prelim tokens. PAS requires no additional forward passes and can be computed on the fly during inference. Exploiting this previously overlooked signal, PAS achieves state-of-the-art object-hallucination detection across multiple models and datasets, enabling real-time filtering and intervention. --- ### Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation Authors: Mohamad Amin Mohamadi, Tianhao Wang, Zhiyuan Li Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11500v3 Modern language models fail a fundamental requirement of trustworthy intelligence: knowing when not to answer. Despite achieving impressive accuracy on benchmarks, these models produce confident hallucinations, even when wrong answers carry catastrophic consequences. Our evaluations on GSM8K, MedQA and GPQA show frontier models almost never abstain despite explicit warnings of severe penalties, suggesting that prompts cannot override training that rewards any answer over no answer. As a remedy, we propose Reinforced Hesitation (RH): a modification to Reinforcement Learning from Verifiable Rewards (RLVR) to use ternary rewards (+1 correct, 0 abstention, -$λ$ error) instead of binary. Controlled experiments on logic puzzles reveal that varying $λ$ produces distinct models along a Pareto frontier, where each training penalty yields the optimal model for its corresponding risk regime: low penalties produce aggressive answerers, high penalties conservative abstainers. The same frontier holds on MATH Levels 4--5 and on medical QA, where it transfers to an unseen dataset. We then introduce two inference strategies that exploit trained abstention as a coordination signal: cascading routes queries through models with decreasing risk tolerance, while self-cascading re-queries the same model on abstention. Both outperform majority voting with lower computational cost. These results establish abstention as a first-class training objective that transforms ``I don't know'' from failure into a coordination signal, enabling models to earn trust through calibrated honesty about their limits. --- ### Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation Authors: Mohamad Amin Mohamadi, Tianhao Wang, Zhiyuan Li Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11500v1 Modern language models fail a fundamental requirement of trustworthy intelligence: knowing when not to answer. Despite achieving impressive accuracy on benchmarks, these models produce confident hallucinations, even when wrong answers carry catastrophic consequences. Our evaluations on GSM8K, MedQA and GPQA show frontier models almost never abstain despite explicit warnings of severe penalties, suggesting that prompts cannot override training that rewards any answer over no answer. As a remedy, we propose Reinforced Hesitation (RH): a modification to Reinforcement Learning from Verifiable Rewards (RLVR) to use ternary rewards (+1 correct, 0 abstention, -$λ$ error) instead of binary. Controlled experiments on logic puzzles reveal that varying $λ$ produces distinct models along a Pareto frontier, where each training penalty yields the optimal model for its corresponding risk regime: low penalties produce aggressive answerers, high penalties conservative abstainers. We then introduce two inference strategies that exploit trained abstention as a coordination signal: cascading routes queries through models with decreasing risk tolerance, while self-cascading re-queries the same model on abstention. Both outperform majority voting with lower computational cost. These results establish abstention as a first-class training objective that transforms ``I don't know'' from failure into a coordination signal, enabling models to earn trust through calibrated honesty about their limits. --- ### Learning and Testing Convex Functions Authors: Renato Ferreira Pinto, Cassandra Marcussen, Elchanan Mossel, Shivam Nadimpalli Published: 2025-11-14 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2511.11498v2 We consider the problems of \emph{learning} and \emph{testing} real-valued convex functions over Gaussian space. Despite the extensive study of function convexity across mathematics, statistics, and computer science, its learnability and testability have largely been examined only in discrete or restricted settings -- typically with respect to the Hamming distance, which is ill-suited for real-valued functions. In contrast, we study these problems in high dimensions under the standard Gaussian measure, assuming sample access to the function and a mild smoothness condition, namely Lipschitzness. A smoothness assumption is natural and, in fact, necessary even in one dimension: without it, convexity cannot be inferred from finitely many samples. As our main results, we give: - Learning Convex Functions: An agnostic proper learning algorithm for Lipschitz convex functions that achieves error $\varepsilon$ using $n^{O(1/\varepsilon^2)}$ samples, together with a complementary lower bound of $n^{\mathrm{poly}(1/\varepsilon)}$ samples in the \emph{correlational statistical query (CSQ)} model. - Testing Convex Functions: A tolerant (two-sided) tester for convexity of Lipschitz functions with the same sample complexity (as a corollary of our learning result), and a one-sided tester (which never rejects convex functions) using $O(\sqrt{n}/\varepsilon)^n$ samples. --- ### Learning and Testing Convex Functions Authors: Renato Ferreira Pinto, Cassandra Marcussen, Elchanan Mossel, Shivam Nadimpalli Published: 2025-11-14 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2511.11498v1 We consider the problems of \emph{learning} and \emph{testing} real-valued convex functions over Gaussian space. Despite the extensive study of function convexity across mathematics, statistics, and computer science, its learnability and testability have largely been examined only in discrete or restricted settings -- typically with respect to the Hamming distance, which is ill-suited for real-valued functions. In contrast, we study these problems in high dimensions under the standard Gaussian measure, assuming sample access to the function and a mild smoothness condition, namely Lipschitzness. A smoothness assumption is natural and, in fact, necessary even in one dimension: without it, convexity cannot be inferred from finitely many samples. As our main results, we give: - Learning Convex Functions: An agnostic proper learning algorithm for Lipschitz convex functions that achieves error $\varepsilon$ using $n^{O(1/\varepsilon^2)}$ samples, together with a complementary lower bound of $n^{\mathrm{poly}(1/\varepsilon)}$ samples in the \emph{correlational statistical query (CSQ)} model. - Testing Convex Functions: A tolerant (two-sided) tester for convexity of Lipschitz functions with the same sample complexity (as a corollary of our learning result), and a one-sided tester (which never rejects convex functions) using $O(\sqrt{n}/\varepsilon)^n$ samples. --- ### Intrinsic Dimension Estimation for Radio Galaxy Zoo using Diffusion Models Authors: Joan Font-Quer Roset, Devina Mohan, Anna Scaife Published: 2025-11-14 Categories: cs.LG, astro-ph.IM, cs.AI arXiv: https://arxiv.org/abs/2511.11490v1 In this work, we estimate the intrinsic dimension (iD) of the Radio Galaxy Zoo (RGZ) dataset using a score-based diffusion model. We examine how the iD estimates vary as a function of Bayesian neural network (BNN) energy scores, which measure how similar the radio sources are to the MiraBest subset of the RGZ dataset. We find that out-of-distribution sources exhibit higher iD values, and that the overall iD for RGZ exceeds those typically reported for natural image datasets. Furthermore, we analyse how iD varies across Fanaroff-Riley (FR) morphological classes and as a function of the signal-to-noise ratio (SNR). While no relationship is found between FR I and FR II classes, a weak trend toward higher SNR at lower iD. Future work using the RGZ dataset could make use of the relationship between iD and energy scores to quantitatively study and improve the representations learned by various self-supervised learning algorithms. --- ### Multimodal Posterior Sampling-based Uncertainty in PD-L1 Segmentation from H&E Images Authors: Roman Kinakh, Gonzalo R. Ríos-Muñoz, Arrate Muñoz-Barrutia Published: 2025-11-14 Categories: cs.CV, q-bio.QM arXiv: https://arxiv.org/abs/2511.11486v1 Accurate assessment of PD-L1 expression is critical for guiding immunotherapy, yet current immunohistochemistry (IHC) based methods are resource-intensive. We present nnUNet-B: a Bayesian segmentation framework that infers PD-L1 expression directly from H&E-stained histology images using Multimodal Posterior Sampling (MPS). Built upon nnUNet-v2, our method samples diverse model checkpoints during cyclic training to approximate the posterior, enabling both accurate segmentation and epistemic uncertainty estimation via entropy and standard deviation. Evaluated on a dataset of lung squamous cell carcinoma, our approach achieves competitive performance against established baselines with mean Dice Score and mean IoU of 0.805 and 0.709, respectively, while providing pixel-wise uncertainty maps. Uncertainty estimates show strong correlation with segmentation error, though calibration remains imperfect. These results suggest that uncertainty-aware H&E-based PD-L1 prediction is a promising step toward scalable, interpretable biomarker assessment in clinical workflows. --- ### Data-efficient U-Net for Segmentation of Carbide Microstructures in SEM Images of Steel Alloys Authors: Alinda Ezgi Gerçek, Till Korten, Paul Chekhonin, Maleeha Hassan, Peter Steinbach Published: 2025-11-14 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2511.11485v1 Understanding reactor-pressure-vessel steel microstructure is crucial for predicting mechanical properties, as carbide precipitates both strengthen the alloy and can initiate cracks. In scanning electron microscopy images, gray-value overlap between carbides and matrix makes simple thresholding ineffective. We present a data-efficient segmentation pipeline using a lightweight U-Net (30.7~M parameters) trained on just \textbf{10 annotated scanning electron microscopy images}. Despite limited data, our model achieves a \textbf{Dice-Sørensen coefficient of 0.98}, significantly outperforming the state-of-the-art in the field of metallurgy (classical image analysis: 0.85), while reducing annotation effort by one order of magnitude compared to the state-of-the-art data efficient segmentation model. This approach enables rapid, automated carbide quantification for alloy design and generalizes to other steel types, demonstrating the potential of data-efficient deep learning in reactor-pressure-vessel steel analysis. --- ### ImAgent: A Unified Multimodal Agent Framework for Test-Time Scalable Image Generation Authors: Kaishen Wang, Ruibo Chen, Tong Zheng, Heng Huang Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11483v4 Recent text-to-image (T2I) models have made remarkable progress in generating visually realistic and semantically coherent images. However, they still suffer from randomness and inconsistency with the given prompts, particularly when textual descriptions are vague or underspecified. Existing approaches, such as prompt rewriting, best-of-N sampling, and self-refinement, can mitigate these issues but usually require additional modules and operate independently, hindering test-time scaling efficiency and increasing computational overhead. In this paper, we introduce ImAgent, a training-free unified multimodal agent that integrates reasoning, generation, and self-evaluation within a single framework for efficient test-time scaling. Guided by a policy controller, multiple generation actions dynamically interact and self-organize to enhance image fidelity and semantic alignment without relying on external models. Extensive experiments on image generation and editing tasks demonstrate that ImAgent consistently improves over the backbone and even surpasses other strong baselines where the backbone model fails, highlighting the potential of unified multimodal agents for adaptive and efficient image generation under test-time scaling. --- ### ImAgent: A Unified Multimodal Agent Framework for Test-Time Scalable Image Generation Authors: Kaishen Wang, Ruibo Chen, Tong Zheng, Heng Huang Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11483v1 Recent text-to-image (T2I) models have made remarkable progress in generating visually realistic and semantically coherent images. However, they still suffer from randomness and inconsistency with the given prompts, particularly when textual descriptions are vague or underspecified. Existing approaches, such as prompt rewriting, best-of-N sampling, and self-refinement, can mitigate these issues but usually require additional modules and operate independently, hindering test-time scaling efficiency and increasing computational overhead. In this paper, we introduce ImAgent, a training-free unified multimodal agent that integrates reasoning, generation, and self-evaluation within a single framework for efficient test-time scaling. Guided by a policy controller, multiple generation actions dynamically interact and self-organize to enhance image fidelity and semantic alignment without relying on external models. Extensive experiments on image generation and editing tasks demonstrate that ImAgent consistently improves over the backbone and even surpasses other strong baselines where the backbone model fails, highlighting the potential of unified multimodal agents for adaptive and efficient image generation under test-time scaling. --- ### Inferring response times of perceptual decisions with Poisson variational autoencoders Authors: Hayden R. Johnson, Anastasia N. Krouglova, Hadi Vafaii, Jacob L. Yates, Pedro J. Gonçalves Published: 2025-11-14 Categories: q-bio.NC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.11480v2 Many properties of perceptual decision making are well-modeled by deep neural networks. However, such architectures typically treat decisions as instantaneous readouts, overlooking the temporal dynamics of the decision process. We present an image-computable model of perceptual decision making in which choices and response times arise from efficient sensory encoding and Bayesian decoding of neural spiking activity. We use a Poisson variational autoencoder to learn unsupervised representations of visual stimuli in a population of rate-coded neurons, modeled as independent homogeneous Poisson processes. A task-optimized decoder then continually infers an approximate posterior over actions conditioned on incoming spiking activity. Combining these components with an entropy-based stopping rule yields a principled and image-computable model of perceptual decisions capable of generating trial-by-trial patterns of choices and response times. Applied to MNIST digit classification, the model reproduces key empirical signatures of perceptual decision making, including stochastic variability, right-skewed response time distributions, logarithmic scaling of response times with the number of alternatives (Hick's law), and speed-accuracy trade-offs. --- ### Inferring response times of perceptual decisions with Poisson variational autoencoders Authors: Hayden R. Johnson, Anastasia N. Krouglova, Hadi Vafaii, Jacob L. Yates, Pedro J. Gonçalves Published: 2025-11-14 Categories: q-bio.NC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2511.11480v1 Many properties of perceptual decision making are well-modeled by deep neural networks. However, such architectures typically treat decisions as instantaneous readouts, overlooking the temporal dynamics of the decision process. We present an image-computable model of perceptual decision making in which choices and response times arise from efficient sensory encoding and Bayesian decoding of neural spiking activity. We use a Poisson variational autoencoder to learn unsupervised representations of visual stimuli in a population of rate-coded neurons, modeled as independent homogeneous Poisson processes. A task-optimized decoder then continually infers an approximate posterior over actions conditioned on incoming spiking activity. Combining these components with an entropy-based stopping rule yields a principled and image-computable model of perceptual decisions capable of generating trial-by-trial patterns of choices and response times. Applied to MNIST digit classification, the model reproduces key empirical signatures of perceptual decision making, including stochastic variability, right-skewed response time distributions, logarithmic scaling of response times with the number of alternatives (Hick's law), and speed-accuracy trade-offs. --- ### Rethinking Progression of Memory State in Robotic Manipulation: An Object-Centric Perspective Authors: Nhat Chung, Taisei Hanyu, Toan Nguyen, Huy Le, Frederick Bumgarner, Duy Minh Ho Nguyen, Khoa Vo, Kashu Yamazaki, Chase Rainwater, Tung Kieu, Anh Nguyen, Ngan Le Published: 2025-11-14 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2511.11478v3 As embodied agents operate in increasingly complex environments, the ability to perceive, track, and reason about individual object instances over time becomes essential, especially in tasks requiring sequenced interactions with visually similar objects. In these non-Markovian settings, key decision cues are often hidden in object-specific histories rather than the current scene. Without persistent memory of prior interactions (what has been interacted with, where it has been, or how it has changed) visuomotor policies may fail, repeat past actions, or overlook completed ones. To surface this challenge, we introduce LIBERO-Mem, a non-Markovian task suite for stress-testing robotic manipulation under object-level partial observability. It combines short- and long-horizon object tracking with temporally sequenced subgoals, requiring reasoning beyond the current frame. However, vision-language-action (VLA) models often struggle in such settings, with token scaling quickly becoming intractable even for tasks spanning just a few hundred frames. We propose Embodied-SlotSSM, a slot-centric VLA framework built for temporal scalability. It maintains spatio-temporally consistent slot identities and leverages them through two mechanisms: (1) slot-state-space modeling for reconstructing short-term history, and (2) a relational encoder to align the input tokens with action decoding. Together, these components enable temporally grounded, context-aware action prediction. Experiments show Embodied-SlotSSM's baseline performance on LIBERO-Mem and general tasks, offering a scalable solution for non-Markovian reasoning in object-centric robotic policies. --- ### Rethinking Progression of Memory State in Robotic Manipulation: An Object-Centric Perspective Authors: Nhat Chung, Taisei Hanyu, Toan Nguyen, Huy Le, Frederick Bumgarner, Duy Minh Ho Nguyen, Khoa Vo, Kashu Yamazaki, Chase Rainwater, Tung Kieu, Anh Nguyen, Ngan Le Published: 2025-11-14 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2511.11478v1 As embodied agents operate in increasingly complex environments, the ability to perceive, track, and reason about individual object instances over time becomes essential, especially in tasks requiring sequenced interactions with visually similar objects. In these non-Markovian settings, key decision cues are often hidden in object-specific histories rather than the current scene. Without persistent memory of prior interactions (what has been interacted with, where it has been, or how it has changed) visuomotor policies may fail, repeat past actions, or overlook completed ones. To surface this challenge, we introduce LIBERO-Mem, a non-Markovian task suite for stress-testing robotic manipulation under object-level partial observability. It combines short- and long-horizon object tracking with temporally sequenced subgoals, requiring reasoning beyond the current frame. However, vision-language-action (VLA) models often struggle in such settings, with token scaling quickly becoming intractable even for tasks spanning just a few hundred frames. We propose Embodied-SlotSSM, a slot-centric VLA framework built for temporal scalability. It maintains spatio-temporally consistent slot identities and leverages them through two mechanisms: (1) slot-state-space modeling for reconstructing short-term history, and (2) a relational encoder to align the input tokens with action decoding. Together, these components enable temporally grounded, context-aware action prediction. Experiments show Embodied-SlotSSM's baseline performance on LIBERO-Mem and general tasks, offering a scalable solution for non-Markovian reasoning in object-centric robotic policies. --- ### Context-aware Adaptive Visualizations for Critical Decision Making Authors: Angela Lopez-Cardona, Mireia Masias Bruns, Nuwan T. Attygalle, Sebastian Idesis, Matteo Salvatori, Konstantinos Raftopoulos, Konstantinos Oikonomou, Saravanakumar Duraisamy, Parvin Emami, Nacera Latreche, Alaa Eddine Anis Sahraoui, Michalis Vakallelis et al. Published: 2025-11-14 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2511.11476v1 Effective decision-making often relies on timely insights from complex visual data. While Information Visualization (InfoVis) dashboards can support this process, they rarely adapt to users' cognitive state, and less so in real time. We present Symbiotik, an intelligent, context-aware adaptive visualization system that leverages neurophysiological signals to estimate mental workload (MWL) and dynamically adapt visual dashboards using reinforcement learning (RL). Through a user study with 120 participants and three visualization types, we demonstrate that our approach improves task performance and engagement. Symbiotik offers a scalable, real-time adaptation architecture, and a validated methodology for neuroadaptive user interfaces. --- ### Proactive Hearing Assistants that Isolate Egocentric Conversations Authors: Guilin Hu, Malek Itani, Tuochao Chen, Shyamnath Gollakota Published: 2025-11-14 Categories: cs.CL, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2511.11473v1 We introduce proactive hearing assistants that automatically identify and separate the wearer's conversation partners, without requiring explicit prompts. Our system operates on egocentric binaural audio and uses the wearer's self-speech as an anchor, leveraging turn-taking behavior and dialogue dynamics to infer conversational partners and suppress others. To enable real-time, on-device operation, we propose a dual-model architecture: a lightweight streaming model runs every 12.5 ms for low-latency extraction of the conversation partners, while a slower model runs less frequently to capture longer-range conversational dynamics. Results on real-world 2- and 3-speaker conversation test sets, collected with binaural egocentric hardware from 11 participants totaling 6.8 hours, show generalization in identifying and isolating conversational partners in multi-conversation settings. Our work marks a step toward hearing assistants that adapt proactively to conversational dynamics and engagement. More information can be found on our website: https://proactivehearing.cs.washington.edu/ --- ### Quantifying and Improving Adaptivity in Conformal Prediction through Input Transformations Authors: Sooyong Jang, Insup Lee Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11472v2 Conformal prediction constructs a set of labels instead of a single point prediction, while providing a probabilistic coverage guarantee. Beyond the coverage guarantee, adaptiveness to example difficulty is an important property. It means that the method should produce larger prediction sets for more difficult examples, and smaller ones for easier examples. Existing evaluation methods for adaptiveness typically analyze coverage rate violation or average set size across bins of examples grouped by difficulty. However, these approaches often suffer from imbalanced binning, which can lead to inaccurate estimates of coverage or set size. To address this issue, we propose a binning method that leverages input transformations to sort examples by difficulty, followed by uniform-mass binning. Building on this binning, we introduce two metrics to better evaluate adaptiveness. These metrics provide more reliable estimates of coverage rate violation and average set size due to balanced binning, leading to more accurate adaptivity assessment. Through experiments, we demonstrate that our proposed metric correlates more strongly with the desired adaptiveness property compared to existing ones. Furthermore, motivated by our findings, we propose a new adaptive prediction set algorithm that groups examples by estimated difficulty and applies group-conditional conformal prediction. This allows us to determine appropriate thresholds for each group. Experimental results on both (a) an Image Classification (ImageNet) (b) a medical task (visual acuity prediction) show that our method outperforms existing approaches according to the new metrics. --- ### Quantifying and Improving Adaptivity in Conformal Prediction through Input Transformations Authors: Sooyong Jang, Insup Lee Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11472v1 Conformal prediction constructs a set of labels instead of a single point prediction, while providing a probabilistic coverage guarantee. Beyond the coverage guarantee, adaptiveness to example difficulty is an important property. It means that the method should produce larger prediction sets for more difficult examples, and smaller ones for easier examples. Existing evaluation methods for adaptiveness typically analyze coverage rate violation or average set size across bins of examples grouped by difficulty. However, these approaches often suffer from imbalanced binning, which can lead to inaccurate estimates of coverage or set size. To address this issue, we propose a binning method that leverages input transformations to sort examples by difficulty, followed by uniform-mass binning. Building on this binning, we introduce two metrics to better evaluate adaptiveness. These metrics provide more reliable estimates of coverage rate violation and average set size due to balanced binning, leading to more accurate adaptivity assessment. Through experiments, we demonstrate that our proposed metric correlates more strongly with the desired adaptiveness property compared to existing ones. Furthermore, motivated by our findings, we propose a new adaptive prediction set algorithm that groups examples by estimated difficulty and applies group-conditional conformal prediction. This allows us to determine appropriate thresholds for each group. Experimental results on both (a) an Image Classification (ImageNet) (b) a medical task (visual acuity prediction) show that our method outperforms existing approaches according to the new metrics. --- ### Sat2RealCity: Geometry-Aware and Appearance-Controllable 3D Urban Generation from Satellite Imagery Authors: Xinliang Wang, Yijie Kang, Zhenyu Wu, Yifeng Shi Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11470v3 3D urban generation from satellite imagery is an important task for scalable digital twins and real-world simulation environments. Existing approaches primarily rely on scene-level generation paradigms, which often require large-scale 3D city assets and struggle with controllability, geographic alignment, and realistic appearance grounding in real-world urban environments. To address these limitations, we present Sat2RealCity, a grounded urban generation framework that leverages object-level 3D generative priors for scalable city synthesis from satellite imagery. Our framework decomposes cities into geographically grounded building entities, enabling the reuse of pretrained object-level 3D generative priors while preserving real-world spatial structures. Supported by our constructed BuildVerse3D dataset, (1) we introduce an OpenStreetMap (OSM)-guided spatial grounding strategy to inject geospatial constraints into the 3D generation process; (2) we design an appearance-guided controllable generation mechanism for realistic architectural appearance and regional style consistency; and (3) we construct an MLLM-powered semantic pipeline for regional appearance understanding and semantic-aware appearance synthesis. Extensive experiments demonstrate that Sat2RealCity achieves strong geographic alignment, regional stylistic consistency, and plausible urban asset synthesis compared with existing urban generation and 3D asset generation approaches. --- ### Sat2RealCity: Geometry-Aware and Appearance-Controllable 3D Urban Generation from Satellite Imagery Authors: Yijie Kang, Xinliang Wang, Zhenyu Wu, Yifeng Shi, Hailong Zhu Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11470v1 Recent advances in generative modeling have substantially enhanced 3D urban generation, enabling applications in digital twins, virtual cities, and large-scale simulations. However, existing methods face two key challenges: (1) the need for large-scale 3D city assets for supervised training, which are difficult and costly to obtain, and (2) reliance on semantic or height maps, which are used exclusively for generating buildings in virtual worlds and lack connection to real-world appearance, limiting the realism and generalizability of generated cities. To address these limitations, we propose Sat2RealCity, a geometry-aware and appearance-controllable framework for 3D urban generation from real-world satellite imagery. Unlike previous city-level generation methods, Sat2RealCity builds generation upon individual building entities, enabling the use of rich priors and pretrained knowledge from 3D object generation while substantially reducing dependence on large-scale 3D city assets. Specifically, (1) we introduce the OSM-based spatial priors strategy to achieve interpretable geometric generation from spatial topology to building instances; (2) we design an appearance-guided controllable modeling mechanism for fine-grained appearance realism and style control; and (3) we construct an MLLM-powered semantic-guided generation pipeline, bridging semantic interpretation and geometric reconstruction. Extensive quantitative and qualitative experiments demonstrate that Sat2RealCity significantly surpasses existing baselines in structural consistency and appearance realism, establishing a strong foundation for real-world aligned 3D urban content creation. The code will be released soon. --- ### Benchmarking Visual LLMs Resilience to Unanswerable Questions on Visually Rich Documents Authors: Davide Napolitano, Luca Cagliero, Fabrizio Battiloro Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11468v1 The evolution of Visual Large Language Models (VLLMs) has revolutionized the automatic understanding of Visually Rich Documents (VRDs), which contain both textual and visual elements. Although VLLMs excel in Visual Question Answering (VQA) on multi-page VRDs, their ability to detect unanswerable questions is still an open research question. Our research delves into the robustness of the VLLMs to plausible yet unanswerable questions, i.e., questions that appear valid but cannot be answered due to subtle corruptions caused by swaps between related concepts or plausible question formulations. Corruptions are generated by replacing the original natural language entities with other ones of the same type, belonging to different document elements, and in different layout positions or pages of the related document. To this end, we present VRD-UQA (VISUALLY RICH DOCUMENT UNANSWERABLE QUESTION ANSWERING), a benchmark for evaluating VLLMs' resilience to plausible yet unanswerable questions across multiple dimensions. It automatically alters the questions of existing VQA datasets consisting of multi-page VRDs, verifies their unanswerability using a VLLM-as-a-judge approach, and then thoroughly evaluates VLLMs' performance. Experiments, run on 12 models, analyze: (1) The VLLMs' accuracy in detecting unanswerable questions at both page and document levels; (2) The effect of different types of corruption (NLP entity, document element, layout); (3) The effectiveness of different knowledge injection strategies based on in-context learning (OCR, multi-page selection, or the possibility of unanswerability). Our findings reveal VLLMs' limitations and demonstrate that VRD-UQA can serve as an evaluation framework for developing resilient document VQA systems. --- ### Non-Euclidean SGD for Structured Optimization: Unified Analysis and Improved Rates Authors: Dmitry Kovalev, Ekaterina Borodich Published: 2025-11-14 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2511.11466v2 Recently, several instances of non-Euclidean SGD, including SignSGD, Lion, and Muon, have attracted significant interest from the optimization community due to their practical success in training deep neural networks. Consequently, a number of works have attempted to explain this success by developing theoretical convergence analyses. Unfortunately, these results cannot properly justify the superior performance of these methods, as they could not beat the convergence rate of vanilla Euclidean SGD. We resolve this important open problem by developing a new unified convergence analysis under the structured smoothness and gradient noise assumption. In particular, our results indicate that non-Euclidean SGD (i) can exploit the sparsity or low-rank structure of the upper bounds on the Hessian and gradient noise, (ii) can provably benefit from popular algorithmic tools such as extrapolation or momentum variance reduction, and (iii) can match the state-of-the-art convergence rates of adaptive and more complex optimization algorithms such as AdaGrad and Shampoo. --- ### Non-Euclidean SGD for Structured Optimization: Unified Analysis and Improved Rates Authors: Dmitry Kovalev, Ekaterina Borodich Published: 2025-11-14 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2511.11466v1 Recently, several instances of non-Euclidean SGD, including SignSGD, Lion, and Muon, have attracted significant interest from the optimization community due to their practical success in training deep neural networks. Consequently, a number of works have attempted to explain this success by developing theoretical convergence analyses. Unfortunately, these results cannot properly justify the superior performance of these methods, as they could not beat the convergence rate of vanilla Euclidean SGD. We resolve this important open problem by developing a new unified convergence analysis under the structured smoothness and gradient noise assumption. In particular, our results indicate that non-Euclidean SGD (i) can exploit the sparsity or low-rank structure of the upper bounds on the Hessian and gradient noise, (ii) can provably benefit from popular algorithmic tools such as extrapolation or momentum variance reduction, and (iii) can match the state-of-the-art convergence rates of adaptive and more complex optimization algorithms such as AdaGrad and Shampoo. --- ### Adaptive Intrusion Detection for Evolving RPL IoT Attacks Using Incremental Learning Authors: Sumeyye Bas, Kiymet Kaya, Elif Ak, Sule Gunduz Oguducu Published: 2025-11-14 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2511.11464v1 The routing protocol for low-power and lossy networks (RPL) has become the de facto routing standard for resource-constrained IoT systems, but its lightweight design exposes critical vulnerabilities to a wide range of routing-layer attacks such as hello flood, decreased rank, and version number manipulation. Traditional countermeasures, including protocol-level modifications and machine learning classifiers, can achieve high accuracy against known threats, yet they fail when confronted with novel or zero-day attacks unless fully retrained, an approach that is impractical for dynamic IoT environments. In this paper, we investigate incremental learning as a practical and adaptive strategy for intrusion detection in RPL-based networks. We systematically evaluate five model families, including ensemble models and deep learning models. Our analysis highlights that incremental learning not only restores detection performance on new attack classes but also mitigates catastrophic forgetting of previously learned threats, all while reducing training time compared to full retraining. By combining five diverse models with attack-specific analysis, forgetting behavior, and time efficiency, this study provides systematic evidence that incremental learning offers a scalable pathway to maintain resilient intrusion detection in evolving RPL-based IoT networks. --- ### MoCap2Radar: A Spatiotemporal Transformer for Synthesizing Micro-Doppler Radar Signatures from Motion Capture Authors: Kevin Chen, Kenneth W. Parker, Anish Arora Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11462v1 We present a pure machine learning process for synthesizing radar spectrograms from Motion-Capture (MoCap) data. We formulate MoCap-to-spectrogram translation as a windowed sequence-to-sequence task using a transformer-based model that jointly captures spatial relations among MoCap markers and temporal dynamics across frames. Real-world experiments show that the proposed approach produces visually and quantitatively plausible doppler radar spectrograms and achieves good generalizability. Ablation experiments show that the learned model includes both the ability to convert multi-part motion into doppler signatures and an understanding of the spatial relations between different parts of the human body. The result is an interesting example of using transformers for time-series signal processing. It is especially applicable to edge computing and Internet of Things (IoT) radars. It also suggests the ability to augment scarce radar datasets using more abundant MoCap data for training higher-level applications. Finally, it requires far less computation than physics-based methods for generating radar data. --- ### Epistemic Error Decomposition for Multi-step Time Series Forecasting: Rethinking Bias-Variance in Recursive and Direct Strategies Authors: Riku Green, Huw Day, Zahraa S. Abdallah, Telmo M. Silva Filho Published: 2025-11-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.11461v1 Multi-step forecasting is often described through a simple rule of thumb: recursive strategies are said to have high bias and low variance, while direct strategies are said to have low bias and high variance. We revisit this belief by decomposing the expected multi-step forecast error into three parts: irreducible noise, a structural approximation gap, and an estimation-variance term. For linear predictors we show that the structural gap is identically zero for any dataset. For nonlinear predictors, however, the repeated composition used in recursion can increase model expressivity, making the structural gap depend on both the model and the data. We further show that the estimation variance of the recursive strategy at any horizon can be written as the one-step variance multiplied by a Jacobian-based amplification factor that measures how sensitive the composed predictor is to parameter error. This perspective explains when recursive forecasting may simultaneously have lower bias and higher variance than direct forecasting. Experiments with multilayer perceptrons on the ETTm1 dataset confirm these findings. The results offer practical guidance for choosing between recursive and direct strategies based on model nonlinearity and noise characteristics, rather than relying on traditional bias-variance intuition. --- ### Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing Classification Authors: Qinghao Gao, Jiahui Qu, Wenqian Dong Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11460v3 Multimodal remote sensing classification often suffers from missing modalities caused by sensor failures and environmental interference, leading to severe performance degradation. In this work, we rethink missing-modality learning from a conditional computation perspective and investigate whether Mixture-of-Experts (MoE) models can inherently adapt to diverse modality-missing scenarios. We first conduct a systematic study of representative MoE paradigms under various missing-modality settings, revealing both their potential and limitations. Building on these insights, we propose a Missing-aware Mixture-of-LoRAs (MaMOL), a parameter-efficient MoE framework that unifies multiple modality-missing cases within a single model. MaMOL introduces a dual-routing mechanism to decouple modality-invariant shared experts and modality-aware dynamic experts, enabling automatic expert activation conditioned on available modalities. Extensive experiments on multiple remote sensing benchmarks demonstrate that MaMOL significantly improves robustness and generalization under diverse missing-modality scenarios with minimal computational overhead. Transfer experiments on natural image datasets further validate its scalability and cross-domain applicability. --- ### Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing Classification Authors: Qinghao Gao, Jianhai Qu, Yunsong Li, Weiqiang Dong Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11460v1 Multimodal classification in remote sensing often suffers from missing modalities caused by environmental interference, sensor failures, or atmospheric effects, which severely degrade classification performance. Existing two-stage adaptation methods are computationally expensive and assume complete multimodal data during training, limiting their generalization to real-world incompleteness. To overcome these issues, we propose a Missing-aware Mixture-of-Loras (MaMOL) framework that reformulates modality missing as a multi-task learning problem. MaMOL introduces a dual-routing mechanism: a task-oriented dynamic router that adaptively activates experts for different missing patterns, and a modality-specific-shared static router that maintains stable cross-modal knowledge sharing. Unlike prior methods that train separate networks for each missing configuration, MaMOL achieves parameter-efficient adaptation via lightweight expert updates and shared expert reuse. Experiments on multiple remote sensing benchmarks demonstrate superior robustness and generalization under varying missing rates, with minimal computational overhead. Moreover, transfer experiments on natural image datasets validate its scalability and cross-domain applicability, highlighting MaMOL as a general and efficient solution for incomplete multimodal learning. --- ### FairReweighing: Density Estimation-Based Reweighing Framework for Improving Separation in Fair Regression Authors: Xiaoyin Xi, Zhe Yu Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11459v1 There has been a prevalence of applying AI software in both high-stakes public-sector and industrial contexts. However, the lack of transparency has raised concerns about whether these data-informed AI software decisions secure fairness against people of all racial, gender, or age groups. Despite extensive research on emerging fairness-aware AI software, up to now most efforts to solve this issue have been dedicated to binary classification tasks. Fairness in regression is relatively underexplored. In this work, we adopted a mutual information-based metric to assess separation violations. The metric is also extended so that it can be directly applied to both classification and regression problems with both binary and continuous sensitive attributes. Inspired by the Reweighing algorithm in fair classification, we proposed a FairReweighing pre-processing algorithm based on density estimation to ensure that the learned model satisfies the separation criterion. Theoretically, we show that the proposed FairReweighing algorithm can guarantee separation in the training data under a data independence assumption. Empirically, on both synthetic and real-world data, we show that FairReweighing outperforms existing state-of-the-art regression fairness solutions in terms of improving separation while maintaining high accuracy. --- ### Synergy vs. Noise: Performance-Guided Multimodal Fusion For Biochemical Recurrence-Free Survival in Prostate Cancer Authors: Seth Alain Chang, Muhammad Mueez Amjad, Noorul Wahab, Ethar Alzaid, Nasir Rajpoot, Adam Shephard Published: 2025-11-14 Categories: q-bio.QM, cs.CV, cs.LG, eess.IV arXiv: https://arxiv.org/abs/2511.11452v1 Multimodal deep learning (MDL) has emerged as a transformative approach in computational pathology. By integrating complementary information from multiple data sources, MDL models have demonstrated superior predictive performance across diverse clinical tasks compared to unimodal models. However, the assumption that combining modalities inherently improves performance remains largely unexamined. We hypothesise that multimodal gains depend critically on the predictive quality of individual modalities, and that integrating weak modalities may introduce noise rather than complementary information. We test this hypothesis on a prostate cancer dataset with histopathology, radiology, and clinical data to predict time-to-biochemical recurrence. Our results confirm that combining high-performing modalities yield superior performance compared to unimodal approaches. However, integrating a poor-performing modality with other higher-performing modalities degrades predictive accuracy. These findings demonstrate that multimodal benefit requires selective, performance-guided integration rather than indiscriminate modality combination, with implications for MDL design across computational pathology and medical imaging. --- ### VoxTell: Free-Text Promptable Universal 3D Medical Image Segmentation Authors: Maximilian Rokuss, Moritz Langenberg, Yannick Kirchhoff, Fabian Isensee, Benjamin Hamm, Constantin Ulrich, Sebastian Regnery, Lukas Bauer, Efthimios Katsigiannopulos, Tobias Norajitra, Klaus Maier-Hein Published: 2025-11-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.11450v1 We introduce VoxTell, a vision-language model for text-prompted volumetric medical image segmentation. It maps free-form descriptions, from single words to full clinical sentences, to 3D masks. Trained on 62K+ CT, MRI, and PET volumes spanning over 1K anatomical and pathological classes, VoxTell uses multi-stage vision-language fusion across decoder layers to align textual and visual features at multiple scales. It achieves state-of-the-art zero-shot performance across modalities on unseen datasets, excelling on familiar concepts while generalizing to related unseen classes. Extensive experiments further demonstrate strong cross-modality transfer, robustness to linguistic variations and clinical language, as well as accurate instance-specific segmentation from real-world text. Code is available at: https://www.github.com/MIC-DKFZ/VoxTell --- ### DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference Authors: Farhana Amin, Sabiha Afroz, Kanchon Gharami, Mona Moghadampanah, Dimitrios S. Nikolopoulos Published: 2025-11-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.11446v1 Diffusion models produce high quality images but inference is costly due to many denoising steps and heavy matrix operations. We present DiffPro, a post-training, hardware-faithful framework that works with the exact integer kernels used in deployment and jointly tunes timesteps and per-layer precision in Diffusion Transformers (DiTs) to reduce latency and memory without any training. DiffPro combines three parts: a manifold-aware sensitivity metric to allocate weight bits, dynamic activation quantization to stabilize activations across timesteps, and a budgeted timestep selector guided by teacher-student drift. In experiments DiffPro achieves up to 6.25x model compression, fifty percent fewer timesteps, and 2.8x faster inference with Delta FID <= 10 on standard benchmarks, demonstrating practical efficiency gains. DiffPro unifies step reduction and precision planning into a single budgeted deployable plan for real-time energy-aware diffusion inference. --- ### Synthetic Stimuli, Real Gains: Rethinking VLM Fine-Tuning Through Fully Controlled Data Generation Authors: Massimo Rizzoli, Simone Alghisi, Seyed Mahed Mousavi, Giuseppe Riccardi Published: 2025-11-14 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2511.11440v3 Performance gains of Vision Language Models (VLMs) obtained by fine-tuning are generally based on ad hoc data collection and annotation of real-world scenes. Despite the improvements, this process is often prone to biases, errors, and distribution imbalance, resulting in overfitting and imbalanced performance. Although a few studies have explored synthetic data generation, they typically lack control over data distribution and annotation quality. In this work, we re-evaluate the potential of model fine-tuning by exploring a fully controlled data generation and annotation pipeline, obtaining bias-free data with balanced distribution and clean annotations. Using the spatial reasoning task of identifying the absolute position of an object as a use case, we fine-tune state-of-the-art VLMs and conduct exhaustive evaluations on both synthetic and real-world benchmarks, including transferability to real-world scenes. Our experiments reveal two key findings: 1) fine-tuning on balanced data yields uniform performance across the visual scene and mitigates common biases with as few as 130 samples; and 2) fine-tuning on synthetic stimuli improves performance by 13% on real-world data (COCO), outperforming models fine-tuned on the full COCO train set. --- ### From Synthetic Scenes to Real Performance: Enhancing Spatial Reasoning in VLMs Authors: Massimo Rizzoli, Simone Alghisi, Seyed Mahed Mousavi, Giuseppe Riccardi Published: 2025-11-14 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2511.11440v1 Fine-tuning Vision-Language Models (VLMs) is a common strategy to improve performance following an ad-hoc data collection and annotation of real-world scenes. However, this process is often prone to biases, errors, and distribution imbalance, resulting in overfitting and imbalanced performance. Although a few studies have tried to address this problem by generating synthetic data, they lacked control over distribution bias and annotation quality. To address these challenges, we redesign the fine-tuning process in two ways. First, we control the generation of data and its annotations, ensuring it is free from bias, distribution imbalance, and annotation errors. We automatically construct the dataset by comprehensively sampling objects' attributes, including color, shape, size, and position within the scene. Secondly, using this annotated dataset, we fine-tune state-of-the-art VLMs and assess performance transferability to real-world data on the absolute position task. We conduct exhaustive evaluations on both synthetic and real-world benchmarks. Our experiments reveal two key findings: 1) fine-tuning on balanced synthetic data yields uniform performance across the visual scene and mitigates common biases; and 2) fine-tuning on synthetic stimuli significantly improves performance on real-world data (COCO), outperforming models fine-tuned in the matched setting. --- ### Retrofit: Continual Learning with Controlled Forgetting for Binary Security Detection and Analysis Authors: Yiling He, Junchi Lei, Hongyu She, Shuo Shao, Xinran Zheng, Yiping Liu, Zhan Qin, Lorenzo Cavallaro Published: 2025-11-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.11439v3 Binary security has increasingly relied on deep learning to reason about malware behavior and program semantics. However, the performance often degrades as threat landscapes evolve and code representations shift. While continual learning (CL) offers a natural solution through sequential updates, most existing approaches rely on data replay or unconstrained updates, limiting their applicability and effectiveness in data-sensitive security environments. We propose RETROFIT, which regulates knowledge retention and adaptation with controlled forgetting at each update, without requiring historical data. Our key idea is to consolidate previously trained and newly fine-tuned models, serving as teachers of legacy and emergent knowledge, through retrospective-free parameter merging. Forgetting control is achieved by 1) constraining parameter changes to low-rank and sparse subspaces for approximate orthogonality, and 2) employing a confidence-guided arbitration mechanism to dynamically aggregate knowledge from both teachers. Our evaluation on two representative applications demonstrates that RETROFIT consistently mitigates forgetting while maintaining adaptability. In malware detection under temporal drift, it substantially improves the retention score, from 20.2% to 38.6% over CL baselines, and exceeds the oracle upper bound on new data. In binary summarization across decompilation levels, where analyzing stripped binaries is especially challenging, RETROFIT achieves over 2x the BLEU score of transfer learning used in prior work and surpasses all baselines in cross-representation generalization. --- ### Retrofit: Continual Learning with Bounded Forgetting for Security Applications Authors: Yiling He, Junchi Lei, Hongyu She, Shuo Shao, Xinran Zheng, Yiping Liu, Zhan Qin, Lorenzo Cavallaro Published: 2025-11-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.11439v1 Modern security analytics are increasingly powered by deep learning models, but their performance often degrades as threat landscapes evolve and data representations shift. While continual learning (CL) offers a promising paradigm to maintain model effectiveness, many approaches rely on full retraining or data replay, which are infeasible in data-sensitive environments. Moreover, existing methods remain inadequate for security-critical scenarios, facing two coupled challenges in knowledge transfer: preserving prior knowledge without old data and integrating new knowledge with minimal interference. We propose RETROFIT, a data retrospective-free continual learning method that achieves bounded forgetting for effective knowledge transfer. Our key idea is to consolidate previously trained and newly fine-tuned models, serving as teachers of old and new knowledge, through parameter-level merging that eliminates the need for historical data. To mitigate interference, we apply low-rank and sparse updates that confine parameter changes to independent subspaces, while a knowledge arbitration dynamically balances the teacher contributions guided by model confidence. Our evaluation on two representative applications demonstrates that RETROFIT consistently mitigates forgetting while maintaining adaptability. In malware detection under temporal drift, it substantially improves the retention score, from 20.2% to 38.6% over CL baselines, and exceeds the oracle upper bound on new data. In binary summarization across decompilation levels, where analyzing stripped binaries is especially challenging, RETROFIT achieves around twice the BLEU score of transfer learning used in prior work and surpasses all baselines in cross-representation generalization. --- ### VP-Bench: A Comprehensive Benchmark for Visual Prompting in Multimodal Large Language Models Authors: Mingjie Xu, Jinpeng Chen, Yuzhi Zhao, Jason Chun Lok Li, Yue Qiu, Zekang Du, Mengyang Wu, Pingping Zhang, Kun Li, Hongzheng Yang, Wenao Ma, Jiaheng Wei et al. Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11438v1 Multimodal large language models (MLLMs) have enabled a wide range of advanced vision-language applications, including fine-grained object recognition and contextual understanding. When querying specific regions or objects in an image, human users naturally use "visual prompts" (VPs), such as bounding boxes, to provide reference. However, no existing benchmark systematically evaluates the ability of MLLMs to interpret such VPs. This gap leaves it unclear whether current MLLMs can effectively recognize VPs, an intuitive prompting method for humans, and use them to solve problems. To address this limitation, we introduce VP-Bench, a benchmark for assessing MLLMs' capability in VP perception and utilization. VP-Bench employs a two-stage evaluation framework: Stage 1 examines models' ability to perceive VPs in natural scenes, using 30k visualized prompts spanning eight shapes and 355 attribute combinations. Stage 2 investigates the impact of VPs on downstream tasks, measuring their effectiveness in real-world problem-solving scenarios. Using VP-Bench, we evaluate 28 MLLMs, including proprietary systems (e.g., GPT-4o) and open-source models (e.g., InternVL3 and Qwen2.5-VL), and provide a comprehensive analysis of factors that affect VP understanding, such as variations in VP attributes, question arrangement, and model scale. VP-Bench establishes a new reference framework for studying how MLLMs comprehend and resolve grounded referring questions. --- ### Hi-DREAM: Brain-Inspired Hierarchical Diffusion for fMRI-to-Image Reconstruction via ROI Encoder and VisuAl Mapping Authors: Guowei Zhang, Yun Zhao, Kai Sun, Moein Khajehnejad, Adeel Razi, Dinh Phung, Levin Kuhlmann Published: 2025-11-14 Categories: cs.CV, cs.HC arXiv: https://arxiv.org/abs/2511.11437v2 Reconstructing natural images from fMRI requires bridging neural activity with both the structural and semantic representations used by modern generative models. Existing diffusion-based decoders often condition on a single global fMRI embedding, which limits their ability to exploit the hierarchical organization of the visual cortex and makes the contribution of different visual areas difficult to inspect. We propose Hi-DREAM, a brain-inspired hierarchical diffusion framework that structures fMRI conditioning according to early, middle, and late visual Regions of Interest (ROI) streams. A ROI adapter converts these streams into a multi-scale cortical pyramid, and a lightweight ROI-conditioned ControlNet injects the resulting anatomy-aware priors into matched U-Net depths during denoising. Experiments on the Natural Scenes Dataset (NSD) show that Hi-DREAM achieves state-of-the-art high-level semantic reconstruction while retaining strong low-level structure. Further ablation and attribution analyses show that the proposed hierarchy-aware conditioning is effective, and that different ROI streams provide complementary, inspectable contributions to reconstruction. --- ### Hi-DREAM: Brain Inspired Hierarchical Diffusion for fMRI Reconstruction via ROI Encoder and visuAl Mapping Authors: Guowei Zhang, Yun Zhao, Moein Khajehnejad, Adeel Razi, Levin Kuhlmann Published: 2025-11-14 Categories: cs.CV, cs.HC arXiv: https://arxiv.org/abs/2511.11437v1 Mapping human brain activity to natural images offers a new window into vision and cognition, yet current diffusion-based decoders face a core difficulty: most condition directly on fMRI features without analyzing how visual information is organized across the cortex. This overlooks the brain's hierarchical processing and blurs the roles of early, middle, and late visual areas. We propose Hi-DREAM, a brain-inspired conditional diffusion framework that makes the cortical organization explicit. A region-of-interest (ROI) adapter groups fMRI into early/mid/late streams and converts them into a multi-scale cortical pyramid aligned with the U-Net depth (shallow scales preserve layout and edges; deeper scales emphasize objects and semantics). A lightweight, depth-matched ControlNet injects these scale-specific hints during denoising. The result is an efficient and interpretable decoder in which each signal plays a brain-like role, allowing the model not only to reconstruct images but also to illuminate functional contributions of different visual areas. Experiments on the Natural Scenes Dataset (NSD) show that Hi-DREAM attains state-of-the-art performance on high-level semantic metrics while maintaining competitive low-level fidelity. These findings suggest that structuring conditioning by cortical hierarchy is a powerful alternative to purely data-driven embeddings and provides a useful lens for studying the visual cortex. --- ### Unsupervised Motion-Compensated Decomposition for Cardiac MRI Reconstruction via Neural Representation Authors: Xuanyu Tian, Lixuan Chen, Qing Wu, Xiao Wang, Jie Feng, Yuyao Zhang, Hongjiang Wei Published: 2025-11-14 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2511.11436v2 Cardiac magnetic resonance (CMR) imaging is widely used to characterize cardiac morphology and function. To accelerate CMR imaging, various methods have been proposed to recover high-quality spatiotemporal CMR images from highly undersampled k-t space data. However, current CMR reconstruction techniques either fail to achieve satisfactory image quality or are restricted by the scarcity of ground truth data, leading to limited applicability in clinical scenarios. In this work, we proposed MoCo-INR, a new unsupervised method that integrates implicit neural representations (INR) with the conventional motion-compensated (MoCo) framework. Using explicit motion modeling and the continuous prior of INRs, MoCo-INR can produce accurate cardiac motion decomposition and high-quality CMR reconstruction. Furthermore, we introduce a new INR network architecture tailored to the CMR problem, which significantly stabilizes model optimization. Experiments on retrospective (simulated) datasets demonstrate the superiority of MoCo-INR over state-of-the-art methods, achieving fast convergence and fine-detailed reconstructions at ultra-high acceleration factors (e.g., 20x in VISTA sampling). Additionally, evaluations on prospective (real-acquired) free-breathing CMR scans highlight the clinical practicality of MoCo-INR for real-time imaging. Several ablation studies further confirm the effectiveness of the critical components of MoCo-INR. --- ### Unsupervised Motion-Compensated Decomposition for Cardiac MRI Reconstruction via Neural Representation Authors: Xuanyu Tian, Lixuan Chen, Qing Wu, Xiao Wang, Jie Feng, Yuyao Zhang, Hongjiang Wei Published: 2025-11-14 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2511.11436v1 Cardiac magnetic resonance (CMR) imaging is widely used to characterize cardiac morphology and function. To accelerate CMR imaging, various methods have been proposed to recover high-quality spatiotemporal CMR images from highly undersampled k-t space data. However, current CMR reconstruction techniques either fail to achieve satisfactory image quality or are restricted by the scarcity of ground truth data, leading to limited applicability in clinical scenarios. In this work, we proposed MoCo-INR, a new unsupervised method that integrates implicit neural representations (INR) with the conventional motion-compensated (MoCo) framework. Using explicit motion modeling and the continuous prior of INRs, MoCo-INR can produce accurate cardiac motion decomposition and high-quality CMR reconstruction. Furthermore, we introduce a new INR network architecture tailored to the CMR problem, which significantly stabilizes model optimization. Experiments on retrospective (simulated) datasets demonstrate the superiority of MoCo-INR over state-of-the-art methods, achieving fast convergence and fine-detailed reconstructions at ultra-high acceleration factors (e.g., 20x in VISTA sampling). Additionally, evaluations on prospective (real-acquired) free-breathing CMR scans highlight the clinical practicality of MoCo-INR for real-time imaging. Several ablation studies further confirm the effectiveness of the critical components of MoCo-INR. --- ### The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion Models Authors: Maria-Teresa De Rosa Palmini, Eva Cetinic Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11435v3 The ambiguity between generalization and memorization in TTI diffusion models becomes pronounced when prompts invoke culturally shared visual references, a phenomenon we term multimodal iconicity. These are instances in which images and texts reflect established cultural associations, such as when a title recalls a familiar artwork or film scene. Such cases challenge existing approaches to evaluating memorization, as they define a setting in which instance-level memorization and culturally grounded generalization are structurally intertwined. To address this challenge, we propose an evaluation framework to assess a model's ability to remain culturally grounded without relying on visual replication. Specifically, we introduce the Cultural Reference Transformation (CRT) metric, which separates two dimensions of model behavior: Recognition, whether a model evokes a reference, from Realization, how it depicts it through replication or reinterpretation. We evaluate five diffusion models on 767 Wikidata-derived cultural references, covering both still and moving imagery, and find differences in how they respond to multimodal iconicity: some show weaker recognition, while others rely more heavily on replication. To assess linguistic sensitivity, we conduct prompt perturbation experiments using synonym substitutions and literal image descriptions, finding that models often reproduce iconic visual structures even when textual cues are altered. Finally, we find that cultural reference recognition correlates not only with training data frequency, but also textual uniqueness, reference popularity, and creation date. Our findings show that the behavior of diffusion models in culturally iconic settings cannot be reduced to simple reproduction, but depends on how references are recognized and realized, advancing evaluation beyond simple text-image matching toward richer contextual understanding. --- ### The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion Models Authors: Maria-Teresa De Rosa Palmini, Eva Cetinic Published: 2025-11-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2511.11435v1 Our work addresses the ambiguity between generalization and memorization in text-to-image diffusion models, focusing on a specific case we term multimodal iconicity. This refers to instances where images and texts evoke culturally shared associations, such as when a title recalls a familiar artwork or film scene. While prior research on memorization and unlearning emphasizes forgetting, we examine what is remembered and how, focusing on the balance between recognizing cultural references and reproducing them. We introduce an evaluation framework that separates recognition, whether a model identifies a reference, from realization, how it depicts it through replication or reinterpretation, quantified through measures capturing both dimensions. By evaluating five diffusion models across 767 Wikidata-derived cultural references spanning static and dynamic imagery, we show that our framework distinguishes replication from transformation more effectively than existing similarity-based methods. To assess linguistic sensitivity, we conduct prompt perturbation experiments using synonym substitutions and literal image descriptions, finding that models often reproduce iconic visual structures even when textual cues are altered. Finally, our analysis shows that cultural alignment correlates not only with training data frequency, but also textual uniqueness, reference popularity, and creation date. Our work reveals that the value of diffusion models lies not only in what they reproduce but in how they transform and recontextualize cultural knowledge, advancing evaluation beyond simple text-image matching toward richer contextual understanding. --- ### WEAVE: Unleashing and Benchmarking the In-context Interleaved Comprehension and Generation Authors: Wei Chow, Jiachun Pan, Yongyuan Liang, Mingze Zhou, Xue Song, Liyu Jia, Saining Zhang, Siliang Tang, Juncheng Li, Fengda Zhang, Weijia Wu, Hanwang Zhang et al. Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11434v1 Recent advances in unified multimodal models (UMMs) have enabled impressive progress in visual comprehension and generation. However, existing datasets and benchmarks focus primarily on single-turn interactions, failing to capture the multi-turn, context-dependent nature of real-world image creation and editing. To address this gap, we present WEAVE, the first suite for in-context interleaved cross-modality comprehension and generation. Our suite consists of two complementary parts. WEAVE-100k is a large-scale dataset of 100K interleaved samples spanning over 370K dialogue turns and 500K images, covering comprehension, editing, and generation tasks that require reasoning over historical context. WEAVEBench is a human-annotated benchmark with 100 tasks based on 480 images, featuring a hybrid VLM judger evaluation framework based on both the reference image and the combination of the original image with editing instructions that assesses models' abilities in multi-turn generation, visual memory, and world-knowledge reasoning across diverse domains. Experiments demonstrate that training on WEAVE-100k enables vision comprehension, image editing, and comprehension-generation collaboration capabilities. Furthermore, it facilitates UMMs to develop emergent visual-memory capabilities, while extensive evaluations on WEAVEBench expose the persistent limitations and challenges of current approaches in multi-turn, context-aware image generation and editing. We believe WEAVE provides a view and foundation for studying in-context interleaved comprehension and generation for multi-modal community. --- ### Comprehension of Multilingual Expressions Referring to Target Objects in Visual Inputs Authors: Francisco Nogueira, Alexandre Bernardino, Bruno Martins Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11427v2 Referring Expression Comprehension (REC) requires models to localize objects in images based on different types of natural language descriptions. Even with significant progress, research on the area remains predominantly English-centric, despite increasing global deployment demands. This work addresses multilingual REC through two main contributions. First, we construct a unified multilingual dataset spanning 10 languages, by systematically expanding 12 existing English REC benchmarks through machine translation and context-based translation enhancement. Second, we introduce an attention-anchored efficient neural architecture that uses a multilingual SigLIP2 encoder. Our attention-based approach generates coarse spatial anchors from attention distributions, which are subsequently refined through learned residuals. Experimental evaluation demonstrates competitive performance on standard benchmarks despite the use of a relatively small model. Multilingual evaluation shows consistent capabilities across languages, establishing the practical feasibility of efficient multilingual visual grounding systems. --- ### Comprehension of Multilingual Expressions Referring to Target Objects in Visual Inputs Authors: Francisco Nogueira, Alexandre Bernardino, Bruno Martins Published: 2025-11-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.11427v1 Referring Expression Comprehension (REC) requires models to localize objects in images based on natural language descriptions. Research on the area remains predominantly English-centric, despite increasing global deployment demands. This work addresses multilingual REC through two main contributions. First, we construct a unified multilingual dataset spanning 10 languages, by systematically expanding 12 existing English REC benchmarks through machine translation and context-based translation enhancement. The resulting dataset comprises approximately 8 million multilingual referring expressions across 177,620 images, with 336,882 annotated objects. Second, we introduce an attention-anchored neural architecture that uses multilingual SigLIP2 encoders. Our attention-based approach generates coarse spatial anchors from attention distributions, which are subsequently refined through learned residuals. Experimental evaluation demonstrates competitive performance on standard benchmarks, e.g. achieving 86.9% accuracy at IoU@50 on RefCOCO aggregate multilingual evaluation, compared to an English-only result of 91.3%. Multilingual evaluation shows consistent capabilities across languages, establishing the practical feasibility of multilingual visual grounding systems. The dataset and model are available at $\href{https://multilingual.franreno.com}{multilingual.franreno.com}$. ---