# Research Papers — 2025-11 · bundle page 166 of 197 > Full-text markdown of papers #8251–8300 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/166 [← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/165.md) [Next bundle →](https://aipod.dev/papers/archive/2025-11/page/167.md) ### Nowcast3D: Reliable precipitation nowcasting via gray-box learning Authors: Huaguan Chen, Wei Han, Haofei Sun, Ning Lin, Xingtao Song, Yunfan Yang, Jie Tian, Yang Liu, Ji-Rong Wen, Xiaoye Zhang, Xueshun Shen, Hao Sun Published: 2025-11-06 Categories: cs.LG, physics.ao-ph arXiv: https://arxiv.org/abs/2511.04659v1 Extreme precipitation nowcasting demands high spatiotemporal fidelity and extended lead times, yet existing approaches remain limited. Numerical Weather Prediction (NWP) and its deep-learning emulations are too slow and coarse for rapidly evolving convection, while extrapolation and purely data-driven models suffer from error accumulation and excessive smoothing. Hybrid 2D radar-based methods discard crucial vertical information, preventing accurate reconstruction of height-dependent dynamics. We introduce a gray-box, fully three-dimensional nowcasting framework that directly processes volumetric radar reflectivity and couples physically constrained neural operators with datadriven learning. The model learns vertically varying 3D advection fields under a conservative advection operator, parameterizes spatially varying diffusion, and introduces a Brownian-motion--inspired stochastic term to represent unresolved motions. A residual branch captures small-scale convective initiation and microphysical variability, while a diffusion-based stochastic module estimates uncertainty. The framework achieves more accurate forecasts up to three-hour lead time across precipitation regimes and ranked first in 57\% of cases in a blind evaluation by 160 meteorologists. By restoring full 3D dynamics with physical consistency, it offers a scalable and robust pathway for skillful and reliable nowcasting of extreme precipitation. --- ### Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcuts Authors: Ellis Brown, Jihan Yang, Shusheng Yang, Rob Fergus, Saining Xie Published: 2025-11-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.04655v2 Multimodal LLMs can answer many questions in vision-centric benchmarks without looking at the image, using linguistic priors and skewed answer distributions. Blind evaluation shows what a model already answers this way, but not what it could learn from the test set's own regularities. Extending partial-input auditing (e.g., hypothesis-only baselines in natural language inference), we argue that benchmark designers should "train on the test set": probe the artifact they release for exploitable patterns. Our Test-set Stress-Test (TsT) cross-validates a text-only Qwen2-7B on the test set's questions and answer options, yielding a benchmark-level score and a per-question bias score s(x); a random forest on hand-crafted features adds a fast, interpretable audit. On the template-based VSI-Bench and CV-Bench, the held-out score is 17.9 and 13.1 points above the model's own zero-shot score. On MMMU it learns little, even though GPT-4o correctly answers 52.3% of its multiple-choice questions without the image, so blind success there reflects pretrained knowledge rather than learnable test-set patterns. Iterative Bias Pruning (IBP) removes the questions with the highest s(x) and re-diagnoses; on VSI-Bench it widens a fine-tuned model's vision-blind gap more than random removal at the same rate. We also release VSI-Bench-Debiased, which lowers a fine-tuned model's blind score from 44.7 to 32.0 while its vision score falls only from 57.1 to 48.7. --- ### Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcuts Authors: Ellis Brown, Jihan Yang, Shusheng Yang, Rob Fergus, Saining Xie Published: 2025-11-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.04655v1 Robust benchmarks are crucial for evaluating Multimodal Large Language Models (MLLMs). Yet we find that models can ace many multimodal benchmarks without strong visual understanding, instead exploiting biases, linguistic priors, and superficial patterns. This is especially problematic for vision-centric benchmarks that are meant to require visual inputs. We adopt a diagnostic principle for benchmark design: if a benchmark can be gamed, it will be. Designers should therefore try to ``game'' their own benchmarks first, using diagnostic and debiasing procedures to systematically identify and mitigate non-visual biases. Effective diagnosis requires directly ``training on the test set'' -- probing the released test set for its intrinsic, exploitable patterns. We operationalize this standard with two components. First, we diagnose benchmark susceptibility using a ``Test-set Stress-Test'' (TsT) methodology. Our primary diagnostic tool involves fine-tuning a powerful Large Language Model via $k$-fold cross-validation on exclusively the non-visual, textual inputs of the test set to reveal shortcut performance and assign each sample a bias score $s(x)$. We complement this with a lightweight Random Forest-based diagnostic operating on hand-crafted features for fast, interpretable auditing. Second, we debias benchmarks by filtering high-bias samples using an ``Iterative Bias Pruning'' (IBP) procedure. Applying this framework to four benchmarks -- VSI-Bench, CV-Bench, MMMU, and VideoMME -- we uncover pervasive non-visual biases. As a case study, we apply our full framework to create VSI-Bench-Debiased, demonstrating reduced non-visual solvability and a wider vision-blind performance gap than the original. --- ### Logit-Entropy Adaptive Stopping Heuristic for Efficient Chain-of-Thought Reasoning Authors: Mohammad Atif Quamar, Mohammad Areeb Published: 2025-11-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.04654v1 Chain-of-Thought (CoT) prompting is a key technique for enabling complex reasoning in large language models. However, generating full, fixed-length rationales is computationally wasteful, inflating both token usage and latency. We introduce LEASH: Logit-Entropy Adaptive Stopping Heuristic, a training-free decoding algorithm that adaptively halts rationale generation. LEASH monitors two intrinsic signals: the slope of token-level entropy and the improvement in the top-logit margin. It terminates the generation once both signals plateau, indicating the model has reached a stable reasoning state. Across four instruction-tuned models on the GSM8K and AQuA-RAT benchmarks, LEASH reduces average token generation by 30--35% and latency by 27%, while incurring a 10 p.p. accuracy drop relative to CoT. LEASH is model-agnostic and requires no additional training or supervision, offering a simple and efficient alternative to CoT decoding. --- ### TT-Prune: Joint Model Pruning and Resource Allocation for Communication-efficient Time-triggered Federated Learning Authors: Xinlu Zhang, Yansha Deng, Toktam Mahmoodi Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04653v1 Federated learning (FL) offers new opportunities in machine learning, particularly in addressing data privacy concerns. In contrast to conventional event-based federated learning, time-triggered federated learning (TT-Fed), as a general form of both asynchronous and synchronous FL, clusters users into different tiers based on fixed time intervals. However, the FL network consists of a growing number of user devices with limited wireless bandwidth, consequently magnifying issues such as stragglers and communication overhead. In this paper, we introduce adaptive model pruning to wireless TT-Fed systems and study the problem of jointly optimizing the pruning ratio and bandwidth allocation to minimize the training loss while ensuring minimal learning latency. To answer this question, we perform convergence analysis on the gradient l_2 norm of the TT-Fed model based on model pruning. Based on the obtained convergence upper bound, a joint optimization problem of pruning ratio and wireless bandwidth is formulated to minimize the model training loss under a given delay threshold. Then, we derive closed-form solutions for wireless bandwidth and pruning ratio using Karush-Kuhn-Tucker(KKT) conditions. The simulation results show that model pruning could reduce the communication cost by 40% while maintaining the model performance at the same level. --- ### Polarization-resolved imaging improves eye tracking Authors: Mantas Žurauskas, Tom Bu, Sanaz Alali, Beyza Kalkanli, Derek Shi, Fernando Alamos, Gauresh Pandit, Christopher Mei, Ali Behrooz, Ramin Mirjalili, Dave Stronks, Alexander Fix et al. Published: 2025-11-06 Categories: cs.CV, physics.optics arXiv: https://arxiv.org/abs/2511.04652v1 Polarization-resolved near-infrared imaging adds a useful optical contrast mechanism to eye tracking by measuring the polarization state of light reflected by ocular tissues in addition to its intensity. In this paper we demonstrate how this contrast can be used to enable eye tracking. Specifically, we demonstrate that a polarization-enabled eye tracking (PET) system composed of a polarization--filter--array camera paired with a linearly polarized near-infrared illuminator can reveal trackable features across the sclera and gaze-informative patterns on the cornea, largely absent in intensity-only images. Across a cohort of 346 participants, convolutional neural network based machine learning models trained on data from PET reduced the median 95th-percentile absolute gaze error by 10--16\% relative to capacity-matched intensity baselines under nominal conditions and in the presence of eyelid occlusions, eye-relief changes, and pupil-size variation. These results link light--tissue polarization effects to practical gains in human--computer interaction and position PET as a simple, robust sensing modality for future wearable devices. --- ### Optimal Inference Schedules for Masked Diffusion Models Authors: Sitan Chen, Kevin Cong, Jerry Li Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04647v2 A major bottleneck of standard auto-regressive large language models is that their inference process is inherently sequential, resulting in very long and costly inference times. To circumvent this, practitioners proposed a class of language models called diffusion language models, of which the masked diffusion model (MDM) is the most successful. The MDM is able to sample tokens out-of-order and, ostensibly, many tokens at once and in parallel. However, there is very limited rigorous understanding of how much parallel sampling these models can perform without noticeable degradation in their sampling performance. Prior work of Li and Cai obtained some preliminary bounds, but these are not tight for many natural classes of distributions. In this work, we give a new, exact characterization of the expected divergence between the true distribution and the sampled distribution, for any distribution and any unmasking schedule for the sampler, showing an elegant connection to the theory of univariate function approximation. By leveraging this connection, we then attain a number of novel lower and upper bounds for this problem. While the connection to function approximation in principle gives the optimal unmasking schedule for any distribution, we show that it is in general impossible to compete with it without strong a priori knowledge of the distribution, even in seemingly benign settings. However, we also demonstrate new upper bounds and new sampling schedules in terms of well-studied information-theoretic properties of the base distribution, namely, its total correlation and dual total correlation, which show that in some natural settings, one can sample in $O(log n)$ steps without any visible loss in performance, where $n$ is the total sequence length. --- ### Optimal Inference Schedules for Masked Diffusion Models Authors: Sitan Chen, Kevin Cong, Jerry Li Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04647v1 A major bottleneck of standard auto-regressive large language models is that their inference process is inherently sequential, resulting in very long and costly inference times. To circumvent this, practitioners proposed a class of language models called diffusion language models, of which the masked diffusion model (MDM) is the most successful. The MDM is able to sample tokens out-of-order and, ostensibly, many tokens at once and in parallel. However, there is very limited rigorous understanding of how much parallel sampling these models can perform without noticeable degradation in their sampling performance. Prior work of Li and Cai obtained some preliminary bounds, but these are not tight for many natural classes of distributions. In this work, we give a new, exact characterization of the expected divergence between the true distribution and the sampled distribution, for any distribution and any unmasking schedule for the sampler, showing an elegant connection to the theory of univariate function approximation. By leveraging this connection, we then attain a number of novel lower and upper bounds for this problem. While the connection to function approximation in principle gives the optimal unmasking schedule for any distribution, we show that it is in general impossible to compete with it without strong a priori knowledge of the distribution, even in seemingly benign settings. However, we also demonstrate new upper bounds and new sampling schedules in terms of well-studied information-theoretic properties of the base distribution, namely, its total correlation and dual total correlation, which show that in some natural settings, one can sample in $O(log n)$ steps without any visible loss in performance, where $n$ is the total sequence length. --- ### DR. WELL: Dynamic Reasoning and Learning with Symbolic World Model for Embodied LLM-Based Multi-Agent Collaboration Authors: Narjes Nourzad, Hanqing Yang, Shiyu Chen, Carlee Joe-Wong Published: 2025-11-06 Categories: cs.AI, cs.CL, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2511.04646v1 Cooperative multi-agent planning requires agents to make joint decisions with partial information and limited communication. Coordination at the trajectory level often fails, as small deviations in timing or movement cascade into conflicts. Symbolic planning mitigates this challenge by raising the level of abstraction and providing a minimal vocabulary of actions that enable synchronization and collective progress. We present DR. WELL, a decentralized neurosymbolic framework for cooperative multi-agent planning. Cooperation unfolds through a two-phase negotiation protocol: agents first propose candidate roles with reasoning and then commit to a joint allocation under consensus and environment constraints. After commitment, each agent independently generates and executes a symbolic plan for its role without revealing detailed trajectories. Plans are grounded in execution outcomes via a shared world model that encodes the current state and is updated as agents act. By reasoning over symbolic plans rather than raw trajectories, DR. WELL avoids brittle step-level alignment and enables higher-level operations that are reusable, synchronizable, and interpretable. Experiments on cooperative block-push tasks show that agents adapt across episodes, with the dynamic world model capturing reusable patterns and improving task completion rates and efficiency. Experiments on cooperative block-push tasks show that our dynamic world model improves task completion and efficiency through negotiation and self-refinement, trading a time overhead for evolving, more efficient collaboration strategies. --- ### When retrieval outperforms generation: Dense evidence retrieval for scalable fake news detection Authors: Alamgir Munir Qazi, John P. McCrae, Jamal Abdul Nasir Published: 2025-11-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.04643v1 The proliferation of misinformation necessitates robust yet computationally efficient fact verification systems. While current state-of-the-art approaches leverage Large Language Models (LLMs) for generating explanatory rationales, these methods face significant computational barriers and hallucination risks in real-world deployments. We present DeReC (Dense Retrieval Classification), a lightweight framework that demonstrates how general-purpose text embeddings can effectively replace autoregressive LLM-based approaches in fact verification tasks. By combining dense retrieval with specialized classification, our system achieves better accuracy while being significantly more efficient. DeReC outperforms explanation-generating LLMs in efficiency, reducing runtime by 95% on RAWFC (23 minutes 36 seconds compared to 454 minutes 12 seconds) and by 92% on LIAR-RAW (134 minutes 14 seconds compared to 1692 minutes 23 seconds), showcasing its effectiveness across varying dataset sizes. On the RAWFC dataset, DeReC achieves an F1 score of 65.58%, surpassing the state-of-the-art method L-Defense (61.20%). Our results demonstrate that carefully engineered retrieval-based systems can match or exceed LLM performance in specialized tasks while being significantly more practical for real-world deployment. --- ### Efficient probabilistic surrogate modeling techniques for partially-observed large-scale dynamical systems Authors: Hans Harder, Abhijeet Vishwasrao, Luca Guastoni, Ricardo Vinuesa, Sebastian Peitz Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04641v1 This paper is concerned with probabilistic techniques for forecasting dynamical systems described by partial differential equations (such as, for example, the Navier-Stokes equations). In particular, it is investigating and comparing various extensions to the flow matching paradigm that reduce the number of sampling steps. In this regard, it compares direct distillation, progressive distillation, adversarial diffusion distillation, Wasserstein GANs and rectified flows. Moreover, experiments are conducted on a set of challenging systems. In particular, we also address the challenge of directly predicting 2D slices of large-scale 3D simulations, paving the way for efficient inflow generation for solvers. --- ### Addressing divergent representations from causal interventions on neural networks Authors: Satchel Grant, Simon Jerome Han, Alexa R. Tartaglini, Christopher Potts Published: 2025-11-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.04638v5 A common approach to mechanistic interpretability is to causally manipulate model representations via targeted interventions in order to understand what those representations encode. Here we ask whether such interventions create out-of-distribution (divergent) representations, and whether this raises concerns about how faithful their resulting explanations are to the target model in its natural state. First, we demonstrate theoretically and empirically that common causal intervention techniques often do shift internal representations away from the natural distribution of the target model. Then, we provide a theoretical analysis of two cases of such divergences: "harmless" divergences that occur in the behavioral null-space of the layer(s) of interest, and "pernicious" divergences that activate hidden network pathways and cause dormant behavioral changes. Finally, in an effort to mitigate the pernicious cases, we apply and modify the Counterfactual Latent (CL) loss from Grant (2025) allowing representations from causal interventions to remain closer to the natural distribution, reducing the likelihood of harmful divergences while preserving the interpretive power of the interventions. Together, these results highlight a path towards more reliable interpretability methods. --- ### Addressing divergent representations from causal interventions on neural networks Authors: Satchel Grant, Simon Jerome Han, Alexa Tartaglini, Christopher Potts Published: 2025-11-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.04638v1 A common approach to mechanistic interpretability is to causally manipulate model representations via targeted interventions in order to understand what those representations encode. Here we ask whether such interventions create out-of-distribution (divergent) representations, and whether this raises concerns about how faithful their resulting explanations are to the target model in its natural state. First, we demonstrate empirically that common causal intervention techniques often do shift internal representations away from the natural distribution of the target model. Then, we provide a theoretical analysis of two classes of such divergences: `harmless' divergences that occur in the null-space of the weights and from covariance within behavioral decision boundaries, and `pernicious' divergences that activate hidden network pathways and cause dormant behavioral changes. Finally, in an effort to mitigate the pernicious cases, we modify the Counterfactual Latent (CL) loss from Grant (2025) that regularizes interventions to remain closer to the natural distributions, reducing the likelihood of harmful divergences while preserving the interpretive power of interventions. Together, these results highlight a path towards more reliable interpretability methods. --- ### NovisVQ: A Streaming Convolutional Neural Network for No-Reference Opinion-Unaware Frame Quality Assessment Authors: Kylie Cancilla, Alexander Moore, Amar Saini, Carmen Carrano Published: 2025-11-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.04628v1 Video quality assessment (VQA) is vital for computer vision tasks, but existing approaches face major limitations: full-reference (FR) metrics require clean reference videos, and most no-reference (NR) models depend on training on costly human opinion labels. Moreover, most opinion-unaware NR methods are image-based, ignoring temporal context critical for video object detection. In this work, we present a scalable, streaming-based VQA model that is both no-reference and opinion-unaware. Our model leverages synthetic degradations of the DAVIS dataset, training a temporal-aware convolutional architecture to predict FR metrics (LPIPS , PSNR, SSIM) directly from degraded video, without references at inference. We show that our streaming approach outperforms our own image-based baseline by generalizing across diverse degradations, underscoring the value of temporal modeling for scalable VQA in real-world vision systems. Additionally, we demonstrate that our model achieves higher correlation with full-reference metrics compared to BRISQUE, a widely-used opinion-aware image quality assessment baseline, validating the effectiveness of our temporal, opinion-unaware approach. --- ### ODE approximation for the Adam algorithm: General and overparametrized setting Authors: Steffen Dereich, Arnulf Jentzen, Sebastian Kassing Published: 2025-11-06 Categories: math.OC, cs.LG, math.PR arXiv: https://arxiv.org/abs/2511.04622v1 The Adam optimizer is currently presumably the most popular optimization method in deep learning. In this article we develop an ODE based method to study the Adam optimizer in a fast-slow scaling regime. For fixed momentum parameters and vanishing step-sizes, we show that the Adam algorithm is an asymptotic pseudo-trajectory of the flow of a particular vector field, which is referred to as the Adam vector field. Leveraging properties of asymptotic pseudo-trajectories, we establish convergence results for the Adam algorithm. In particular, in a very general setting we show that if the Adam algorithm converges, then the limit must be a zero of the Adam vector field, rather than a local minimizer or critical point of the objective function. In contrast, in the overparametrized empirical risk minimization setting, the Adam algorithm is able to locally find the set of minima. Specifically, we show that in a neighborhood of the global minima, the objective function serves as a Lyapunov function for the flow induced by the Adam vector field. As a consequence, if the Adam algorithm enters a neighborhood of the global minima infinitely often, it converges to the set of global minima. --- ### Dynamic causal discovery in Alzheimer's disease through latent pseudotime modelling Authors: Natalia Glazman, Jyoti Mangal, Pedro Borges, Sebastien Ourselin, M. Jorge Cardoso Published: 2025-11-06 Categories: stat.AP, cs.CE, cs.LG arXiv: https://arxiv.org/abs/2511.04619v1 The application of causal discovery to diseases like Alzheimer's (AD) is limited by the static graph assumptions of most methods; such models cannot account for an evolving pathophysiology, modulated by a latent disease pseudotime. We propose to apply an existing latent variable model to real-world AD data, inferring a pseudotime that orders patients along a data-driven disease trajectory independent of chronological age, then learning how causal relationships evolve. Pseudotime outperformed age in predicting diagnosis (AUC 0.82 vs 0.59). Incorporating minimal, disease-agnostic background knowledge substantially improved graph accuracy and orientation. Our framework reveals dynamic interactions between novel (NfL, GFAP) and established AD markers, enabling practical causal discovery despite violated assumptions. --- ### Building Trust in Virtual Immunohistochemistry: Automated Assessment of Image Quality Authors: Tushar Kataria, Shikha Dubey, Mary Bronner, Jolanta Jedrzkiewicz, Ben J. Brintz, Shireen Y. Elhabian, Beatrice S. Knudsen Published: 2025-11-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.04615v1 Deep learning models can generate virtual immunohistochemistry (IHC) stains from hematoxylin and eosin (H&E) images, offering a scalable and low-cost alternative to laboratory IHC. However, reliable evaluation of image quality remains a challenge as current texture- and distribution-based metrics quantify image fidelity rather than the accuracy of IHC staining. Here, we introduce an automated and accuracy grounded framework to determine image quality across sixteen paired or unpaired image translation models. Using color deconvolution, we generate masks of pixels stained brown (i.e., IHC-positive) as predicted by each virtual IHC model. We use the segmented masks of real and virtual IHC to compute stain accuracy metrics (Dice, IoU, Hausdorff distance) that directly quantify correct pixel - level labeling without needing expert manual annotations. Our results demonstrate that conventional image fidelity metrics, including Frechet Inception Distance (FID), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM), correlate poorly with stain accuracy and pathologist assessment. Paired models such as PyramidPix2Pix and AdaptiveNCE achieve the highest stain accuracy, whereas unpaired diffusion- and GAN-based models are less reliable in providing accurate IHC positive pixel labels. Moreover, whole-slide images (WSI) reveal performance declines that are invisible in patch-based evaluations, emphasizing the need for WSI-level benchmarks. Together, this framework defines a reproducible approach for assessing the quality of virtual IHC models, a critical step to accelerate translation towards routine use by pathologists. --- ### evomap: A Toolbox for Dynamic Mapping in Python Authors: Maximilian Matthe Published: 2025-11-06 Categories: cs.MS, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2511.04611v1 This paper presents evomap, a Python package for dynamic mapping. Mapping methods are widely used across disciplines to visualize relationships among objects as spatial representations, or maps. However, most existing statistical software supports only static mapping, which captures objects' relationships at a single point in time and lacks tools to analyze how these relationships evolve. evomap fills this gap by implementing the dynamic mapping framework EvoMap, originally proposed by Matthe, Ringel, and Skiera (2023), which adapts traditional static mapping methods for dynamic analyses. The package supports multiple mapping techniques, including variants of Multidimensional Scaling (MDS), Sammon Mapping, and t-distributed Stochastic Neighbor Embedding (t-SNE). It also includes utilities for data preprocessing, exploration, and result evaluation, offering a comprehensive toolkit for dynamic mapping applications. This paper outlines the foundations of static and dynamic mapping, describes the architecture and functionality of evomap, and illustrates its application through an extensive usage example. --- ### PixCLIP: Achieving Fine-grained Visual Language Understanding via Any-granularity Pixel-Text Alignment Learning Authors: Yicheng Xiao, Yu Chen, Haoxuan Ma, Jiale Hong, Caorui Li, Lingxiang Wu, Haiyun Guo, Jinqiao Wang Published: 2025-11-06 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2511.04601v1 While the Contrastive Language-Image Pretraining(CLIP) model has achieved remarkable success in a variety of downstream vison language understanding tasks, enhancing its capability for fine-grained image-text alignment remains an active research focus. To this end, most existing works adopt the strategy of explicitly increasing the granularity of visual information processing, e.g., incorporating visual prompts to guide the model focus on specific local regions within the image. Meanwhile, researches on Multimodal Large Language Models(MLLMs) have demonstrated that training with long and detailed textual descriptions can effectively improve the model's fine-grained vision-language alignment. However, the inherent token length limitation of CLIP's text encoder fundamentally limits CLIP to process more granular textual information embedded in long text sequences. To synergistically leverage the advantages of enhancing both visual and textual content processing granularity, we propose PixCLIP, a novel framework designed to concurrently accommodate visual prompt inputs and process lengthy textual descriptions. Specifically, we first establish an automated annotation pipeline capable of generating pixel-level localized, long-form textual descriptions for images. Utilizing this pipeline, we construct LongGRIT, a high-quality dataset comprising nearly 1.5 million samples. Secondly, we replace CLIP's original text encoder with the LLM and propose a three-branch pixel-text alignment learning framework, facilitating fine-grained alignment between image regions and corresponding textual descriptions at arbitrary granularity. Experiments demonstrate that PixCLIP showcases breakthroughs in pixel-level interaction and handling long-form texts, achieving state-of-the-art performance. --- ### From Global to Local Correlation: Geometric Decomposition of Statistical Inference Authors: Pawel Gajer, Jacques Ravel Published: 2025-11-06 Categories: stat.ME, math.ST, stat.ML arXiv: https://arxiv.org/abs/2511.04599v4 Understanding feature-outcome associations in high-dimensional data remains challenging when relationships vary across subpopulations, yet standard methods assuming global associations miss context-dependent patterns, reducing statistical power and interpretability. We develop a geometric decomposition framework offering two strategies for partitioning inference problems into regional analyses on data-derived Riemannian graphs. Gradient flow decomposition uses path-monotonicity-validated discrete Morse theory to partition samples into gradient flow cells where outcomes exhibit monotonic behavior. Co-monotonicity decomposition utilizes vertex-level coefficients that provide context-dependent versions of the classical Pearson correlation: these coefficients measure edge-based directional concordance between outcome and features, or between feature pairs, defining embeddings of samples into association space. These embeddings induce Riemannian k-NN graphs on which biclustering identifies co-monotonicity cells (coherent regions) and feature modules. This extends naturally to multi-modal integration across multiple feature sets. Both strategies apply independently or jointly, with Bayesian posterior sampling providing credible intervals. --- ### Geometric Decomposition of Statistical Inference through Gradient Flow and Co-Monotonicity Measures Authors: Pawel Gajer, Jacques Ravel Published: 2025-11-06 Categories: stat.ME, math.ST, stat.ML, stat.TH, 62G08, 62H30, 58E05 arXiv: https://arxiv.org/abs/2511.04599v1 Understanding feature-outcome associations in high-dimensional data remains challenging when relationships vary across subpopulations, yet standard methods assuming global associations miss context-dependent patterns, reducing statistical power and interpretability. We develop a geometric decomposition framework offering two strategies for partitioning inference problems into regional analyses on data-derived Riemannian graphs. Gradient flow decomposition uses path-monotonicity-validated discrete Morse theory to partition samples into basins where outcomes exhibit monotonic behavior. Co-monotonicity decomposition leverages association structure: vertex-level coefficients measuring directional concordance between outcome and features, or between feature pairs, define embeddings of samples into association space. These embeddings induce Riemannian k-NN graphs on which biclustering identifies co-monotonicity cells (coherent regions) and feature modules. This extends naturally to multi-modal integration across multiple feature sets. Both strategies apply independently or jointly, with Bayesian posterior sampling providing credible intervals. --- ### Environment Agnostic Goal-Conditioning, A Study of Reward-Free Autonomous Learning Authors: Hampus Åström, Elin Anna Topp, Jacek Malec Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04598v1 In this paper we study how transforming regular reinforcement learning environments into goal-conditioned environments can let agents learn to solve tasks autonomously and reward-free. We show that an agent can learn to solve tasks by selecting its own goals in an environment-agnostic way, at training times comparable to externally guided reinforcement learning. Our method is independent of the underlying off-policy learning algorithm. Since our method is environment-agnostic, the agent does not value any goals higher than others, leading to instability in performance for individual goals. However, in our experiments, we show that the average goal success rate improves and stabilizes. An agent trained with this method can be instructed to seek any observations made in the environment, enabling generic training of agents prior to specific use cases. --- ### UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstruction Authors: Chen Shi, Shaoshuai Shi, Xiaoyang Lyu, Chunyang Liu, Kehua Sheng, Bo Zhang, Li Jiang Published: 2025-11-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2511.04595v1 Feed-forward 3D reconstruction for autonomous driving has advanced rapidly, yet existing methods struggle with the joint challenges of sparse, non-overlapping camera views and complex scene dynamics. We present UniSplat, a general feed-forward framework that learns robust dynamic scene reconstruction through unified latent spatio-temporal fusion. UniSplat constructs a 3D latent scaffold, a structured representation that captures geometric and semantic scene context by leveraging pretrained foundation models. To effectively integrate information across spatial views and temporal frames, we introduce an efficient fusion mechanism that operates directly within the 3D scaffold, enabling consistent spatio-temporal alignment. To ensure complete and detailed reconstructions, we design a dual-branch decoder that generates dynamic-aware Gaussians from the fused scaffold by combining point-anchored refinement with voxel-based generation, and maintain a persistent memory of static Gaussians to enable streaming scene completion beyond current camera coverage. Extensive experiments on real-world datasets demonstrate that UniSplat achieves state-of-the-art performance in novel view synthesis, while providing robust and high-quality renderings even for viewpoints outside the original camera coverage. --- ### Regret Lower Bounds for Decentralized Multi-Agent Stochastic Shortest Path Problems Authors: Utkarsh U. Chavan, Prashant Trivedi, Nandyala Hemachandra Published: 2025-11-06 Categories: cs.LG, cs.MA arXiv: https://arxiv.org/abs/2511.04594v2 Multi-agent systems (MAS) are central to applications such as swarm robotics and traffic routing, where agents must coordinate in a decentralized manner to achieve a common objective. Stochastic Shortest Path (SSP) problems provide a natural framework for modeling decentralized control in such settings. While the problem of learning in SSP has been extensively studied in single-agent settings, the decentralized multi-agent variant remains largely unexplored. In this work, we take a step towards addressing that gap. We study decentralized multi-agent SSPs (Dec-MASSPs) under linear function approximation, where the transition dynamics and costs are represented using linear models. Applying novel symmetry-based arguments, we identify the structure of optimal policies. Our main contribution is the first regret lower bound for this setting based on the construction of hard-to-learn instances for any number of agents, $n$. Our regret lower bound of $Ω(\sqrt{K})$, over $K$ episodes, highlights the inherent learning difficulty in Dec-MASSPs. These insights clarify the learning complexity of decentralized control and can further guide the design of efficient learning algorithms in multi-agent systems. --- ### Regret Lower Bounds for Decentralized Multi-Agent Stochastic Shortest Path Problems Authors: Utkarsh U. Chavan, Prashant Trivedi, Nandyala Hemachandra Published: 2025-11-06 Categories: cs.LG, cs.MA arXiv: https://arxiv.org/abs/2511.04594v1 Multi-agent systems (MAS) are central to applications such as swarm robotics and traffic routing, where agents must coordinate in a decentralized manner to achieve a common objective. Stochastic Shortest Path (SSP) problems provide a natural framework for modeling decentralized control in such settings. While the problem of learning in SSP has been extensively studied in single-agent settings, the decentralized multi-agent variant remains largely unexplored. In this work, we take a step towards addressing that gap. We study decentralized multi-agent SSPs (Dec-MASSPs) under linear function approximation, where the transition dynamics and costs are represented using linear models. Applying novel symmetry-based arguments, we identify the structure of optimal policies. Our main contribution is the first regret lower bound for this setting based on the construction of hard-to-learn instances for any number of agents, $n$. Our regret lower bound of $\Omega(\sqrt{K})$, over $K$ episodes, highlights the inherent learning difficulty in Dec-MASSPs. These insights clarify the learning complexity of decentralized control and can further guide the design of efficient learning algorithms in multi-agent systems. --- ### Neural Computation Without Slots: Steps Towards Biologically Plausible Memory and Attention in Natural and Artificial Intelligence Authors: Shaunak Bhandarkar, James L. McClelland Published: 2025-11-06 Categories: cs.NE, q-bio.NC arXiv: https://arxiv.org/abs/2511.04593v1 Many models used in artificial intelligence and cognitive science rely on multi-element patterns stored in "slots" - dedicated storage locations - in a digital computer. As biological brains likely lack slots, we consider how they might achieve similar functional outcomes without them by building on the neurally-inspired modern Hopfield network (MHN; Krotov & Hopfield, 2021), which stores patterns in the connection weights of an individual neuron. We propose extensions of this approach to increase its biological plausibility as a model of memory and to capture an important advantage of slot-based computation in contemporary language models. For memory, neuroscience research suggests that the weights of overlapping sparse ensembles of neurons, rather than a dedicated individual neuron, are used to store a memory. We introduce the K-winner MHN, extending the approach to ensembles, and find that within a continual learning regime, the ensemble-based MHN exhibits greater retention of older memories, as measured by the graded sensitivity measure d', than a standard (one-neuron) MHN. Next, we consider the powerful use of slot-based memory in contemporary language models. These models use slots to store long sequences of past inputs and their learned encodings, supporting later predictions and allowing error signals to be transported backward in time to adjust weights underlying the learned encodings of these past inputs. Inspired by these models' successes, we show how the MHN can be extended to capture both of these important functional outcomes. Collectively, our modeling approaches constitute steps towards understanding how biologically plausible mechanisms can support computations that have enabled AI systems to capture human-like abilities that no prior models have been able to achieve. --- ### Complexity as Advantage: A Regret-Based Perspective on Emergent Structure Authors: Oshri Naparstek Published: 2025-11-06 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2511.04590v1 We introduce Complexity as Advantage (CAA), a framework that defines the complexity of a system relative to a family of observers. Instead of measuring complexity as an intrinsic property, we evaluate how much predictive regret a system induces for different observers attempting to model it. A system is complex when it is easy for some observers and hard for others, creating an information advantage. We show that this formulation unifies several notions of emergent behavior, including multiscale entropy, predictive information, and observer-dependent structure. The framework suggests that "interesting" systems are those positioned to create differentiated regret across observers, providing a quantitative grounding for why complexity can be functionally valuable. We demonstrate the idea through simple dynamical models and discuss implications for learning, evolution, and artificial agents. --- ### Question the Questions: Auditing Representation in Online Deliberative Processes Authors: Soham De, Lodewijk Gelauff, Ashish Goel, Smitha Milli, Ariel Procaccia, Alice Siu Published: 2025-11-06 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2511.04588v1 A central feature of many deliberative processes, such as citizens' assemblies and deliberative polls, is the opportunity for participants to engage directly with experts. While participants are typically invited to propose questions for expert panels, only a limited number can be selected due to time constraints. This raises the challenge of how to choose a small set of questions that best represent the interests of all participants. We introduce an auditing framework for measuring the level of representation provided by a slate of questions, based on the social choice concept known as justified representation (JR). We present the first algorithms for auditing JR in the general utility setting, with our most efficient algorithm achieving a runtime of $O(mn\log n)$, where $n$ is the number of participants and $m$ is the number of proposed questions. We apply our auditing methods to historical deliberations, comparing the representativeness of (a) the actual questions posed to the expert panel (chosen by a moderator), (b) participants' questions chosen via integer linear programming, (c) summary questions generated by large language models (LLMs). Our results highlight both the promise and current limitations of LLMs in supporting deliberative processes. By integrating our methods into an online deliberation platform that has been used for over hundreds of deliberations across more than 50 countries, we make it easy for practitioners to audit and improve representation in future deliberations. --- ### Are We Asking the Right Questions? On Ambiguity in Natural Language Queries for Tabular Data Analysis Authors: Daniel Gomm, Cornelius Wolff, Madelon Hulsebos Published: 2025-11-06 Categories: cs.AI, cs.CL, cs.DB, cs.HC arXiv: https://arxiv.org/abs/2511.04584v4 Natural language interfaces to tabular data must handle ambiguities inherent to queries. Instead of treating ambiguity as a deficiency, we reframe it as a feature of cooperative interaction where users are intentional about the degree to which they specify queries. We develop a principled framework based on a shared responsibility of query specification between user and system, distinguishing unambiguous and ambiguous cooperative queries, which systems can resolve through reasonable inference, from uncooperative queries that cannot be resolved. Applying the framework to evaluations for tabular question answering and analysis, we analyze queries in 15 datasets, and observe an uncontrolled mixing of query types neither adequate for evaluating a system's accuracy nor for evaluating interpretation capabilities. This conceptualization around cooperation in resolving queries informs how to design and evaluate natural language interfaces for tabular data analysis, for which we distill concrete directions for future research and broader implications. --- ### Are We Asking the Right Questions? On Ambiguity in Natural Language Queries for Tabular Data Analysis Authors: Daniel Gomm, Cornelius Wolff, Madelon Hulsebos Published: 2025-11-06 Categories: cs.AI, cs.CL, cs.DB, cs.HC arXiv: https://arxiv.org/abs/2511.04584v1 Natural language interfaces to tabular data must handle ambiguities inherent to queries. Instead of treating ambiguity as a deficiency, we reframe it as a feature of cooperative interaction, where the responsibility of query specification is shared among the user and the system. We develop a principled framework distinguishing cooperative queries, i.e., queries that yield a resolvable interpretation, from uncooperative queries that cannot be resolved. Applying the framework to evaluations for tabular question answering and analysis, we analyze the queries in 15 popular datasets, and observe an uncontrolled mixing of query types neither adequate for evaluating a system's execution accuracy nor for evaluating interpretation capabilities. Our framework and analysis of queries shifts the perspective from fixing ambiguity to embracing cooperation in resolving queries. This reflection enables more informed design and evaluation for natural language interfaces for tabular data, for which we outline implications and directions for future research. --- ### Jr. AI Scientist and Its Risk Report: Autonomous Scientific Exploration from a Baseline Paper Authors: Atsuyuki Miyai, Mashiro Toyooka, Takashi Otonari, Zaiying Zhao, Kiyoharu Aizawa Published: 2025-11-06 Categories: cs.AI, cs.CL, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.04583v4 Understanding the current capabilities and risks of AI Scientist systems (autoresearch) is essential for ensuring trustworthy and sustainable AI-driven scientific progress while preserving the integrity of the academic ecosystem. To this end, we develop Jr. AI Scientist, a state-of-the-art autonomous AI scientist system that mimics the core research workflow of a novice student researcher: Given the baseline paper from the human mentor, it analyzes its limitations, formulates novel hypotheses for improvement, iteratively experiments until improvements are achieved, and writes a paper with the results. Unlike previous approaches that assume full automation or operate on small-scale code, Jr. AI Scientist follows a well-defined research workflow and leverages modern coding agents to handle complex, multi-file implementations, leading to scientifically valuable contributions. Through our experiments, the Jr. AI Scientist successfully generated new research papers that build upon real NeurIPS, IJCV, and ICLR works by proposing and implementing novel methods. For evaluation, we conducted automated assessments using AI Reviewers, author-led evaluations, and submissions to Agents4Science, a venue dedicated to AI-driven contributions. The findings demonstrate that Jr. AI Scientist generates papers receiving higher review scores by DeepReviewer than existing fully automated systems. Nevertheless, we identify important limitations from the author evaluation and the Agents4Science reviews, indicating the potential risks of directly applying current AI Scientist systems and key challenges for future research. Finally, we comprehensively report various risks identified during development. We believe this study clarifies the current role and limitations of AI Scientist systems, offering insights into the areas that still require human expertise and the risks that may emerge as these systems evolve. --- ### Jr. AI Scientist and Its Risk Report: Autonomous Scientific Exploration from a Baseline Paper Authors: Atsuyuki Miyai, Mashiro Toyooka, Takashi Otonari, Zaiying Zhao, Kiyoharu Aizawa Published: 2025-11-06 Categories: cs.AI, cs.CL, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2511.04583v1 Understanding the current capabilities and risks of AI Scientist systems is essential for ensuring trustworthy and sustainable AI-driven scientific progress while preserving the integrity of the academic ecosystem. To this end, we develop Jr. AI Scientist, a state-of-the-art autonomous AI scientist system that mimics the core research workflow of a novice student researcher: Given the baseline paper from the human mentor, it analyzes its limitations, formulates novel hypotheses for improvement, validates them through rigorous experimentation, and writes a paper with the results. Unlike previous approaches that assume full automation or operate on small-scale code, Jr. AI Scientist follows a well-defined research workflow and leverages modern coding agents to handle complex, multi-file implementations, leading to scientifically valuable contributions. For evaluation, we conducted automated assessments using AI Reviewers, author-led evaluations, and submissions to Agents4Science, a venue dedicated to AI-driven scientific contributions. The findings demonstrate that Jr. AI Scientist generates papers receiving higher review scores than existing fully automated systems. Nevertheless, we identify important limitations from both the author evaluation and the Agents4Science reviews, indicating the potential risks of directly applying current AI Scientist systems and key challenges for future research. Finally, we comprehensively report various risks identified during development. We hope these insights will deepen understanding of current progress and risks in AI Scientist development. --- ### Physics-Informed Neural Networks and Neural Operators for Parametric PDEs Authors: Zhuo Zhang, Xiong Xiong, Sen Zhang, Yuan Zhao, Xi Yang Published: 2025-11-06 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2511.04576v3 PDEs arise ubiquitously in science and engineering, where solutions depend on parameters (physical properties, boundary conditions, geometry). Traditional numerical methods require re-solving the PDE for each parameter, making parameter space exploration prohibitively expensive. Recent machine learning advances, particularly physics-informed neural networks (PINNs) and neural operators, have revolutionized parametric PDE solving by learning solution operators that generalize across parameter spaces. We critically analyze two main paradigms: (1) PINNs, which embed physical laws as soft constraints and excel at inverse problems with sparse data, and (2) neural operators (e.g., DeepONet, Fourier Neural Operator), which learn mappings between infinite-dimensional function spaces and achieve unprecedented generalization. Through comparisons across fluid dynamics, solid mechanics, heat transfer, and electromagnetics, we show neural operators can achieve computational speedups of $10^3$ to $10^5$ times faster than traditional solvers for multi-query scenarios, while maintaining comparable accuracy. We provide practical guidance for method selection, discuss theoretical foundations (universal approximation, convergence), and identify critical open challenges: high-dimensional parameters, complex geometries, and out-of-distribution generalization. This work establishes a unified framework for understanding parametric PDE solvers via operator learning, offering a comprehensive, incrementally updated resource for this rapidly evolving field --- ### Physics-Informed Neural Networks and Neural Operators for Parametric PDEs: A Human-AI Collaborative Analysis Authors: Zhuo Zhang, Xiong Xiong, Sen Zhang, Yuan Zhao, Xi Yang Published: 2025-11-06 Categories: stat.ML, cs.LG, 68T01 arXiv: https://arxiv.org/abs/2511.04576v2 PDEs arise ubiquitously in science and engineering, where solutions depend on parameters (physical properties, boundary conditions, geometry). Traditional numerical methods require re-solving the PDE for each parameter, making parameter space exploration prohibitively expensive. Recent machine learning advances, particularly physics-informed neural networks (PINNs) and neural operators, have revolutionized parametric PDE solving by learning solution operators that generalize across parameter spaces. We critically analyze two main paradigms: (1) PINNs, which embed physical laws as soft constraints and excel at inverse problems with sparse data, and (2) neural operators (e.g., DeepONet, Fourier Neural Operator), which learn mappings between infinite-dimensional function spaces and achieve unprecedented generalization. Through comparisons across fluid dynamics, solid mechanics, heat transfer, and electromagnetics, we show neural operators can achieve computational speedups of $10^3$ to $10^5$ times faster than traditional solvers for multi-query scenarios, while maintaining comparable accuracy. We provide practical guidance for method selection, discuss theoretical foundations (universal approximation, convergence), and identify critical open challenges: high-dimensional parameters, complex geometries, and out-of-distribution generalization. This work establishes a unified framework for understanding parametric PDE solvers via operator learning, offering a comprehensive, incrementally updated resource for this rapidly evolving field --- ### Physics-Informed Neural Networks and Neural Operators for Parametric PDEs: A Human-AI Collaborative Analysis Authors: Zhuo Zhang, Xiong Xiong, Sen Zhang, Yuan Zhao, Xi Yang Published: 2025-11-06 Categories: stat.ML, cs.LG, 68T01 arXiv: https://arxiv.org/abs/2511.04576v1 PDEs arise ubiquitously in science and engineering, where solutions depend on parameters (physical properties, boundary conditions, geometry). Traditional numerical methods require re-solving the PDE for each parameter, making parameter space exploration prohibitively expensive. Recent machine learning advances, particularly physics-informed neural networks (PINNs) and neural operators, have revolutionized parametric PDE solving by learning solution operators that generalize across parameter spaces. We critically analyze two main paradigms: (1) PINNs, which embed physical laws as soft constraints and excel at inverse problems with sparse data, and (2) neural operators (e.g., DeepONet, Fourier Neural Operator), which learn mappings between infinite-dimensional function spaces and achieve unprecedented generalization. Through comparisons across fluid dynamics, solid mechanics, heat transfer, and electromagnetics, we show neural operators can achieve computational speedups of $10^3$ to $10^5$ times faster than traditional solvers for multi-query scenarios, while maintaining comparable accuracy. We provide practical guidance for method selection, discuss theoretical foundations (universal approximation, convergence), and identify critical open challenges: high-dimensional parameters, complex geometries, and out-of-distribution generalization. This work establishes a unified framework for understanding parametric PDE solvers via operator learning, offering a comprehensive, incrementally updated resource for this rapidly evolving field --- ### ARETE: an R package for Automated REtrieval from TExt with large language models Authors: Vasco V. Branco, Jandó Benedek, Lidia Pivovarova, Luís Correia, Pedro Cardoso Published: 2025-11-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2511.04573v1 1. A hard stop for the implementation of rigorous conservation initiatives is our lack of key species data, especially occurrence data. Furthermore, researchers have to contend with an accelerated speed at which new information must be collected and processed due to anthropogenic activity. Publications ranging from scientific papers to gray literature contain this crucial information but their data are often not machine-readable, requiring extensive human work to be retrieved. 2. We present the ARETE R package, an open-source software aiming to automate data extraction of species occurrences powered by large language models, namely using the chatGPT Application Programming Interface. This R package integrates all steps of the data extraction and validation process, from Optical Character Recognition to detection of outliers and output in tabular format. Furthermore, we validate ARETE through systematic comparison between what is modelled and the work of human annotators. 3. We demonstrate the usefulness of the approach by comparing range maps produced using GBIF data and with those automatically extracted for 100 species of spiders. Newly extracted data allowed to expand the known Extent of Occurrence by a mean three orders of magnitude, revealing new areas where the species were found in the past, which mayhave important implications for spatial conservation planning and extinction risk assessments. 4. ARETE allows faster access to hitherto untapped occurrence data, a potential game changer in projects requiring such data. Researchers will be able to better prioritize resources, manually verifying selected species while maintaining automated extraction for the majority. This workflow also allows predicting available bibliographic data during project planning. --- ### Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm Authors: Jingqi Tong, Yurong Mou, Hangcheng Li, Mingzhe Li, Yongzhuo Yang, Ming Zhang, Qiguang Chen, Tianyi Liang, Xiaomeng Hu, Yining Zheng, Xinchi Chen, Jun Zhao et al. Published: 2025-11-06 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2511.04570v2 The "Thinking with Text" and "Thinking with Images" paradigms significantly improve the reasoning abilities of large language models (LLMs) and Vision-Language Models (VLMs). However, these paradigms have inherent limitations. (1) Images capture only single moments and fail to represent dynamic processes or continuous changes, and (2) The separation of text and vision as distinct modalities, which hinders unified multimodal understanding and generation. Therefore, we propose "Thinking with Video", a new paradigm that leverages video generation models such as Sora-2 to use video frames as a unified medium for multimodal reasoning. To support this exploration, we developed the Video Thinking Benchmark (VideoThinkBench), which covers both vision-centric tasks (e.g., Eyeballing Puzzles) and text-centric tasks (e.g., GSM8K and MMMU). Our evaluation on VideoThinkBench establishes Sora-2 as a capable reasoner. On vision-centric tasks, Sora-2 is comparable to state-of-the-art (SOTA) VLMs, and even surpasses GPT-5 by 10% on eyeballing puzzles. On text-centric tasks, Sora-2 achieves 92% accuracy on MATH, and 69.2% accuracy on MMMU. Furthermore, we systematically analyze the source of these abilities. We also find that self-consistency and in-context learning can improve Sora-2's performance. In summary, our findings show that the video generation model is the potential unified multimodal understanding and generation model, positioning "Thinking with Video" as a potential unified multimodal reasoning paradigm. --- ### Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm Authors: Jingqi Tong, Yurong Mou, Hangcheng Li, Mingzhe Li, Yongzhuo Yang, Ming Zhang, Qiguang Chen, Tianyi Liang, Xiaomeng Hu, Yining Zheng, Xinchi Chen, Jun Zhao et al. Published: 2025-11-06 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2511.04570v1 "Thinking with Text" and "Thinking with Images" paradigm significantly improve the reasoning ability of large language models (LLMs) and Vision Language Models (VLMs). However, these paradigms have inherent limitations. (1) Images capture only single moments and fail to represent dynamic processes or continuous changes, and (2) The separation of text and vision as distinct modalities, hindering unified multimodal understanding and generation. To overcome these limitations, we introduce "Thinking with Video", a new paradigm that leverages video generation models, such as Sora-2, to bridge visual and textual reasoning in a unified temporal framework. To support this exploration, we developed the Video Thinking Benchmark (VideoThinkBench). VideoThinkBench encompasses two task categories: (1) vision-centric tasks (e.g., Eyeballing Puzzles), and (2) text-centric tasks (e.g., subsets of GSM8K, MMMU). Our evaluation establishes Sora-2 as a capable reasoner. On vision-centric tasks, Sora-2 is generally comparable to state-of-the-art (SOTA) VLMs, and even surpasses VLMs on several tasks, such as Eyeballing Games. On text-centric tasks, Sora-2 achieves 92% accuracy on MATH, and 75.53% accuracy on MMMU. Furthermore, we systematically analyse the source of these abilities. We also find that self-consistency and in-context learning can improve Sora-2's performance. In summary, our findings demonstrate that the video generation model is the potential unified multimodal understanding and generation model, positions "thinking with video" as a unified multimodal reasoning paradigm. --- ### Riesz Regression As Direct Density Ratio Estimation Authors: Masahiro Kato Published: 2025-11-06 Categories: stat.ML, cs.LG, econ.EM, math.ST, stat.ME arXiv: https://arxiv.org/abs/2511.04568v2 This study clarifies the relationship between Riesz regression (Chernozhukov et al., 2021) and density ratio estimation (DRE) in causal inference problems, such as average treatment effect estimation. We first show that the Riesz representer can be written as a signed density ratio and then demonstrate that the Riesz regression objective coincides with the least-squares importance fitting criterion (Kanamori et al., 2009). Although Riesz regression applies to a broad class of representer estimation problems, this equivalence with DRE allows us to transfer existing DRE results, including convergence rate analyses, generalizations based on Bregman divergence minimization, and regularization techniques for flexible models such as neural networks. --- ### Riesz Regression As Direct Density Ratio Estimation Authors: Masahiro Kato Published: 2025-11-06 Categories: stat.ML, cs.LG, econ.EM, math.ST, stat.ME, stat.TH arXiv: https://arxiv.org/abs/2511.04568v1 Riesz regression has garnered attention as a tool in debiased machine learning for causal and structural parameter estimation (Chernozhukov et al., 2021). This study shows that Riesz regression is closely related to direct density-ratio estimation (DRE) in important cases, including average treat- ment effect (ATE) estimation. Specifically, the idea and objective in Riesz regression coincide with the one in least-squares importance fitting (LSIF, Kanamori et al., 2009) in direct density-ratio estimation. While Riesz regression is general in the sense that it can be applied to Riesz representer estimation in a wide class of problems, the equivalence with DRE allows us to directly import exist- ing results in specific cases, including convergence-rate analyses, the selection of loss functions via Bregman-divergence minimization, and regularization techniques for flexible models, such as neural networks. Conversely, insights about the Riesz representer in debiased machine learning broaden the applications of direct density-ratio estimation methods. This paper consolidates our prior results in Kato (2025a) and Kato (2025b). --- ### Machine Learning for Electron-Scale Turbulence Modeling in W7-X Authors: Ionut-Gabriel Farcas, Don Lawrence Carl Agapito Fernando, Alejandro Banon Navarro, Gabriele Merlo, Frank Jenko Published: 2025-11-06 Categories: physics.plasm-ph, cs.CE, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2511.04567v2 Constructing reduced models for turbulent transport is essential for accelerating profile predictions and enabling many-query tasks such as parameter exploration and design optimization. This work investigates machine-learning-driven reduced models for Electron Temperature Gradient (ETG) turbulence in the Wendelstein 7-X (W7-X) stellarator. We develop physics-guided scaling laws to predict the ETG heat flux at seven radial locations as functions of three key plasma parameters: the normalized electron temperature gradient ($ω_{T_e}$), the ratio of normalized electron temperature and density gradients ($η_e$), and the electron-to-ion temperature ratio ($τ$). The model coefficients are determined through regression combined with an active learning strategy. The procedure initializes the scaling laws using low-cardinality sparse-grid training data and iteratively enriches the training set by selecting maximally informative samples from an existing simulation database. The predictive performance of the models is assessed using out-of-sample datasets comprising more than $393$ points per radial location. Using the coefficients identified at the seven training radial locations, we further derive regression-based parameterizations for the scaling-law coefficients as functions of radial position. The resulting models are then evaluated at three additional radial locations not used during training, including both interpolation and moderate extrapolation cases. Overall, our reduced models demonstrate good predictive performance and achieve accuracy comparable to the original reference simulations, including in interpolation and moderate extrapolation regimes. An important finding is that a single radius-independent model cannot adequately describe ETG transport across the W7-X core, suggesting the presence of geometry-dependent physics not captured by the present formulation. --- ### Machine Learning for Electron-Scale Turbulence Modeling in W7-X Authors: Ionut-Gabriel Farcas, Don Lawrence Carl Agapito Fernando, Alejandro Banon Navarro, Gabriele Merlo, Frank Jenko Published: 2025-11-06 Categories: physics.plasm-ph, cs.CE, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2511.04567v1 Constructing reduced models for turbulent transport is essential for accelerating profile predictions and enabling many-query tasks such as uncertainty quantification, parameter scans, and design optimization. This paper presents machine-learning-driven reduced models for Electron Temperature Gradient (ETG) turbulence in the Wendelstein 7-X (W7-X) stellarator. Each model predicts the ETG heat flux as a function of three plasma parameters: the normalized electron temperature radial gradient ($\omega_{T_e}$), the ratio of normalized electron temperature and density radial gradients ($\eta_e$), and the electron-to-ion temperature ratio ($\tau$). We first construct models across seven radial locations using regression and an active machine-learning-based procedure. This process initializes models using low-cardinality sparse-grid training data and then iteratively refines their training sets by selecting the most informative points from a pre-existing simulation database. We evaluate the prediction capabilities of our models using out-of-sample datasets with over $393$ points per location, and $95\%$ prediction intervals are estimated via bootstrapping to assess prediction uncertainty. We then investigate the construction of generalized reduced models, including a generic, position-independent model, and assess their heat flux prediction capabilities at three additional locations. Our models demonstrate robust performance and predictive accuracy comparable to the original reference simulations, even when applied beyond the training domain. --- ### Uncertainties in Physics-informed Inverse Problems: The Hidden Risk in Scientific AI Authors: Yoh-ichi Mototake, Makoto Sasaki Published: 2025-11-06 Categories: physics.comp-ph, cs.LG arXiv: https://arxiv.org/abs/2511.04564v1 Physics-informed machine learning (PIML) integrates partial differential equations (PDEs) into machine learning models to solve inverse problems, such as estimating coefficient functions (e.g., the Hamiltonian function) that characterize physical systems. This framework enables data-driven understanding and prediction of complex physical phenomena. While coefficient functions in PIML are typically estimated on the basis of predictive performance, physics as a discipline does not rely solely on prediction accuracy to evaluate models. For example, Kepler's heliocentric model was favored owing to small discrepancies in planetary motion, despite its similar predictive accuracy to the geocentric model. This highlights the inherent uncertainties in data-driven model inference and the scientific importance of selecting physically meaningful solutions. In this paper, we propose a framework to quantify and analyze such uncertainties in the estimation of coefficient functions in PIML. We apply our framework to reduced model of magnetohydrodynamics and our framework shows that there are uncertainties, and unique identification is possible with geometric constraints. Finally, we confirm that we can estimate the reduced model uniquely by incorporating these constraints. --- ### BanglaMedQA and BanglaMMedBench: Evaluating Retrieval-Augmented Generation Strategies for Bangla Biomedical Question Answering Authors: Sadia Sultana, Saiyma Sittul Muna, Mosammat Zannatul Samarukh, Ajwad Abrar, Tareque Mohmud Chowdhury Published: 2025-11-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2511.04560v1 Developing accurate biomedical Question Answering (QA) systems in low-resource languages remains a major challenge, limiting equitable access to reliable medical knowledge. This paper introduces BanglaMedQA and BanglaMMedBench, the first large-scale Bangla biomedical Multiple Choice Question (MCQ) datasets designed to evaluate reasoning and retrieval in medical artificial intelligence (AI). The study applies and benchmarks several Retrieval-Augmented Generation (RAG) strategies, including Traditional, Zero-Shot Fallback, Agentic, Iterative Feedback, and Aggregate RAG, combining textbook-based and web retrieval with generative reasoning to improve factual accuracy. A key novelty lies in integrating a Bangla medical textbook corpus through Optical Character Recognition (OCR) and implementing an Agentic RAG pipeline that dynamically selects between retrieval and reasoning strategies. Experimental results show that the Agentic RAG achieved the highest accuracy 89.54% with openai/gpt-oss-120b, outperforming other configurations and demonstrating superior rationale quality. These findings highlight the potential of RAG-based methods to enhance the reliability and accessibility of Bangla medical QA, establishing a foundation for future research in multilingual medical artificial intelligence. --- ### Integrating Temporal and Structural Context in Graph Transformers for Relational Deep Learning Authors: Divyansha Lachi, Mahmoud Mohammadi, Joe Meyer, Vinam Arora, Tom Palczewski, Eva L. Dyer Published: 2025-11-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2511.04557v1 In domains such as healthcare, finance, and e-commerce, the temporal dynamics of relational data emerge from complex interactions-such as those between patients and providers, or users and products across diverse categories. To be broadly useful, models operating on these data must integrate long-range spatial and temporal dependencies across diverse types of entities, while also supporting multiple predictive tasks. However, existing graph models for relational data primarily focus on spatial structure, treating temporal information merely as a filtering constraint to exclude future events rather than a modeling signal, and are typically designed for single-task prediction. To address these gaps, we introduce a temporal subgraph sampler that enhances global context by retrieving nodes beyond the immediate neighborhood to capture temporally relevant relationships. In addition, we propose the Relational Graph Perceiver (RGP), a graph transformer architecture for relational deep learning that leverages a cross-attention-based latent bottleneck to efficiently integrate information from both structural and temporal contexts. This latent bottleneck integrates signals from different node and edge types into a common latent space, enabling the model to build global context across the entire relational system. RGP also incorporates a flexible cross-attention decoder that supports joint learning across tasks with disjoint label spaces within a single model. Experiments on RelBench, SALT, and CTU show that RGP delivers state-of-the-art performance, offering a general and scalable solution for relational deep learning with support for diverse predictive tasks. --- ### Flow Reconstruction from Sparse Measurements in Urban Drainage Networks: An Application and Evaluation of Data-Driven Sparse Sensing Authors: Zihang Ding, Amit Kumar, Imran Md. Azizul Islam, Mila Avellar Montezuma, Ruihang Zhang, Kun Zhang Published: 2025-11-06 Categories: cs.AI, cs.CE arXiv: https://arxiv.org/abs/2511.04556v3 Urbanization and increasingly frequent intense storms are placing stress on urban drainage networks. While dense monitoring of urban drainage networks is desirable, practical constraints in time, budget, and technology hinder its full implementation. How to monitor and predict flow conditions across the entire network under constrained resources is a major challenge. To address this, we utilized and evaluated an established data-driven sparse sensing (DSS) workflow for sensor placement optimization and sewer flow reconstruction in a 77-node urban drainage network. A validated SWMM parameterization was used to construct a spatial basis using singular value decomposition (SVD), select rank-specific layouts using pivoted QR, and define the reconstruction decoder. Applied to 225 held-out simulations combining 25 plausible calibrated parameter sets with 9 rainfall events, a 3-node monitoring layout (4% of the network) achieved a median system-level Nash-Sutcliffe efficiency (NSE) of 0.791 and a 10th percentile of 0.719; all simulations exceeded NSE = 0.700. The pivoted QR-selected sensor sets were benchmarked against reference sensor configurations obtained from Greedy D-optimal and genetic algorithm, matching their performance without requiring iterative layout searches. We further evaluated the framework's robustness by introducing multiplicative Gaussian noise and simulating individual sensor failures, finding that performance was insensitive to noise but varied based on the location of lost monitored nodes. The reconstruction's sensitivity to monitored node loss was consistently associated with energy-weighted modal exposure with a leave-one-node-out Spearman correlation coefficient of 0.781, which was strongly correlated with the lost node's upstream drainage area, mean adjacent circular-conduit diameter, and nodal active flow fraction. --- ### Optimizing Sensor Placement in Urban Storm Sewers: A Data-Driven Sparse Sensing Approach Authors: Zihang Ding, Kun Zhang Published: 2025-11-06 Categories: cs.AI, cs.CE arXiv: https://arxiv.org/abs/2511.04556v1 Urban surface water flooding, triggered by intense rainfall overwhelming drainage systems, is increasingly frequent and widespread. While flood prediction and monitoring in high spatial-temporal resolution are desired, practical constraints in time, budget, and technology hinder its full implementation. How to monitor urban drainage networks and predict flow conditions under constrained resource is a major challenge. This study presents a data-driven sparse sensing (DSS) framework, integrated with EPA-SWMM, to optimize sensor placement and reconstruct peak flowrates in a stormwater system, using the Woodland Avenue catchment in Duluth, Minnesota, as a case study. We utilized a SWMM model to generate a training dataset of peak flowrate profiles across the stormwater network. Furthermore, we applied DSS - leveraging singular value decomposition for dimensionality reduction and QR factorization for sensor allocation - to identify the optimal monitoring nodes based on the simulated training dataset. We then validated the representativeness of these identified monitoring nodes by comparing the DSS-reconstructed peak flowrate profiles with those obtained from SWMM. Three optimally placed sensors among 77 nodes achieved satisfactory reconstruction performance with Nash-Sutcliffe Efficiency (NSE) values of 0.92-0.95 (25th to 75th percentiles). In addition, the model showed good robustness to uncertainty in measurements. Its robustness to sensor failures is location-dependent and improves with the number of sensors deployed. The framework balances computational efficiency and physical interpretability, enabling high-accuracy flow reconstruction with minimal sensors. This DSS framework can be further integrated with predictive models to realize flood early warning and real-time control under limited sensing and monitoring resource. --- ### Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment Authors: Tao Lin, Yilei Zhong, Yuxin Du, Jingjing Zhang, Jiting Liu, Yinxinyu Chen, Encheng Gu, Ziyan Liu, Hongyi Cai, Yanwen Zou, Lixing Zou, Zhaoye Zhou et al. Published: 2025-11-06 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2511.04555v2 Vision-Language-Action (VLA) models have emerged as a powerful framework that unifies perception, language, and control, enabling robots to perform diverse tasks through multimodal understanding. However, current VLA models typically contain massive parameters and rely heavily on large-scale robot data pretraining, leading to high computational costs during training, as well as limited deployability for real-time inference. Moreover, most training paradigms often degrade the perceptual representations of the vision-language backbone, resulting in overfitting and poor generalization to downstream tasks. In this work, we present Evo-1, a lightweight VLA model that reduces computation and improves deployment efficiency, while maintaining strong performance without pretraining on robot data. Evo-1 builds on a native multimodal Vision-Language model (VLM), incorporating a novel cross-modulated diffusion transformer along with an optimized integration module, together forming an effective architecture. We further introduce a two-stage training paradigm that progressively aligns action with perception, preserving the representations of the VLM. Notably, with only 0.77 billion parameters, Evo-1 achieves state-of-the-art results on the Meta-World and RoboTwin suite, surpassing the previous best models by 12.4% and 6.9%, respectively, and also attains a competitive result of 94.8% on LIBERO. In real-world evaluations, Evo-1 attains a 78% success rate with high inference frequency and low memory overhead, outperforming all baseline methods. We release code, data, and model weights to facilitate future research on lightweight and efficient VLA models. --- ### Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment Authors: Tao Lin, Yilei Zhong, Yuxin Du, Jingjing Zhang, Jiting Liu, Yinxinyu Chen, Encheng Gu, Ziyan Liu, Hongyi Cai, Yanwen Zou, Lixing Zou, Zhaoye Zhou et al. Published: 2025-11-06 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2511.04555v1 Vision-Language-Action (VLA) models have emerged as a powerful framework that unifies perception, language, and control, enabling robots to perform diverse tasks through multimodal understanding. However, current VLA models typically contain massive parameters and rely heavily on large-scale robot data pretraining, leading to high computational costs during training, as well as limited deployability for real-time inference. Moreover, most training paradigms often degrade the perceptual representations of the vision-language backbone, resulting in overfitting and poor generalization to downstream tasks. In this work, we present Evo-1, a lightweight VLA model that reduces computation and improves deployment efficiency, while maintaining strong performance without pretraining on robot data. Evo-1 builds on a native multimodal Vision-Language model (VLM), incorporating a novel cross-modulated diffusion transformer along with an optimized integration module, together forming an effective architecture. We further introduce a two-stage training paradigm that progressively aligns action with perception, preserving the representations of the VLM. Notably, with only 0.77 billion parameters, Evo-1 achieves state-of-the-art results on the Meta-World and RoboTwin suite, surpassing the previous best models by 12.4% and 6.9%, respectively, and also attains a competitive result of 94.8% on LIBERO. In real-world evaluations, Evo-1 attains a 78% success rate with high inference frequency and low memory overhead, outperforming all baseline methods. We release code, data, and model weights to facilitate future research on lightweight and efficient VLA models. --- ### Generative Bayesian Filtering and Parameter Learning Authors: Edoardo Marcelli, Sean O'Hagan, Veronika Rockova Published: 2025-11-06 Categories: stat.ME, stat.CO, stat.ML arXiv: https://arxiv.org/abs/2511.04552v1 Generative Bayesian Filtering (GBF) provides a powerful and flexible framework for performing posterior inference in complex nonlinear and non-Gaussian state-space models. Our approach extends Generative Bayesian Computation (GBC) to dynamic settings, enabling recursive posterior inference using simulation-based methods powered by deep neural networks. GBF does not require explicit density evaluations, making it particularly effective when observation or transition distributions are analytically intractable. To address parameter learning, we introduce the Generative-Gibbs sampler, which bypasses explicit density evaluation by iteratively sampling each variable from its implicit full conditional distribution. Such technique is broadly applicable and enables inference in hierarchical Bayesian models with intractable densities, including state-space models. We assess the performance of the proposed methodologies through both simulated and empirical studies, including the estimation of $α$-stable stochastic volatility models. Our findings indicate that GBF significantly outperforms existing likelihood-free approaches in accuracy and robustness when dealing with intractable state-space models. ---