# Research Papers — 2025-12 · bundle page 154 of 175 > Full-text markdown of papers #7651–7700 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2025-12/page/154 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/153.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/155.md) ### Large Language Models for Limited Noisy Data: A Gravitational Wave Identification Study Authors: Yixuan Li, Yuhao Lu, Yang Liu, Liang Li, R. Ruffini, Di Li, Rong-Gen Cai, Xiaoyan Zhu, Wenbin Lin, Yu Wang Published: 2025-12-03 Categories: astro-ph.IM, astro-ph.HE, cs.AI arXiv: https://arxiv.org/abs/2512.04031v1 This work investigates whether large language models (LLMs) offer advantages over traditional neural networks for astronomical data processing, in regimes with non-Gaussian, non-stationary noise and limited labeled samples. Gravitational wave observations provide an suitable test case, using only 90 LIGO events, finetuned LLMs achieve 97.4\% accuracy for identifying signals. Further experiments show that, in contrast to traditional networks that rely on large simulated datasets, additional simulated samples do not improve LLM performance, while scaling studies reveal predictable gains with increasing model size and dataset size. These results indicate that LLMs can extract discriminative structure directly from observational data and provide an efficient assessment for gravitational wave identification. The same strategy may extend to other astronomical domains with similar noise properties, such as radio or pulsar observations. --- ### PSA: Pyramid Sparse Attention for Efficient Video Understanding and Generation Authors: Xiaolong Li, Youping Gu, Xi Lin, Weijie Wang, Bohan Zhuang Published: 2025-12-03 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.04025v1 Attention mechanisms are the core of foundation models, but their quadratic complexity remains a critical bottleneck for scaling. This challenge has driven the development of efficient attention mechanisms, with sparsity emerging as the dominant paradigm. Current methods typically retain or discard entire key-value blocks with binary masks, resulting in substantial information loss under high sparsity. To mitigate this gap, we present Pyramid Sparse Attention (PSA), a versatile module applicable to both video understanding and generation tasks. Instead of binary masking, PSA introduces multi-level pooled KV representations, enabling finer mask granularity. Specifically, each query block dynamically allocates lower pooling levels to critical KV blocks and higher levels to less important ones, creating an informative interpolation between full retention and complete pruning. This design, analogous to fixed-point quantization and classical feature pyramid networks in computer vision, effectively mitigates information loss while preserving computational efficiency under a low compute budget. It works with a native, hardware-friendly kernel that leverages decoupled block-tile design to ensure efficient execution. Across video understanding and generation benchmarks, PSA preserves contextual information and visual fidelity, consistently outperforming or achieving comparable performance over existing sparse attention baselines with superior efficiency-quality trade-offs. Our code and model weights are publicly available at: http://ziplab.co/PSA --- ### C3G: Learning Compact 3D Representations with 2K Gaussians Authors: Honggyu An, Jaewoo Jung, Mungyeom Kim, Chaehyun Kim, Minkyeong Jeon, Jisang Han, Kazumi Fukuda, Takuya Narihira, Hyuna Ko, Junsu Kim, Sunghwan Hong, Yuki Mitsufuji et al. Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.04021v2 Reconstructing and understanding 3D scenes from unposed sparse views in a feed-forward manner remains as a challenging task in 3D computer vision. Recent approaches use per-pixel 3D Gaussian Splatting for reconstruction, followed by a 2D-to-3D feature lifting stage for scene understanding. However, they generate excessive redundant Gaussians, causing high memory overhead and sub-optimal multi-view feature aggregation, leading to degraded novel view synthesis and scene understanding performance. We propose C3G, a novel feed-forward framework that estimates compact 3D Gaussians only at essential spatial locations, minimizing redundancy while enabling effective feature lifting. We introduce learnable tokens that aggregate multi-view features through self-attention to guide Gaussian generation, ensuring each Gaussian integrates relevant visual features across views. We then exploit the learned attention patterns for Gaussian decoding to efficiently lift features. Extensive experiments on pose-free novel view synthesis, 3D open-vocabulary segmentation, and view-invariant feature aggregation demonstrate our approach's effectiveness. Results show that a compact yet geometrically meaningful representation is sufficient for high-quality scene reconstruction and understanding, achieving superior memory efficiency and feature fidelity compared to existing methods. --- ### C3G: Learning Compact 3D Representations with 2K Gaussians Authors: Honggyu An, Jaewoo Jung, Mungyeom Kim, Sunghwan Hong, Chaehyun Kim, Kazumi Fukuda, Minkyeong Jeon, Jisang Han, Takuya Narihira, Hyuna Ko, Junsu Kim, Yuki Mitsufuji et al. Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.04021v1 Reconstructing and understanding 3D scenes from unposed sparse views in a feed-forward manner remains as a challenging task in 3D computer vision. Recent approaches use per-pixel 3D Gaussian Splatting for reconstruction, followed by a 2D-to-3D feature lifting stage for scene understanding. However, they generate excessive redundant Gaussians, causing high memory overhead and sub-optimal multi-view feature aggregation, leading to degraded novel view synthesis and scene understanding performance. We propose C3G, a novel feed-forward framework that estimates compact 3D Gaussians only at essential spatial locations, minimizing redundancy while enabling effective feature lifting. We introduce learnable tokens that aggregate multi-view features through self-attention to guide Gaussian generation, ensuring each Gaussian integrates relevant visual features across views. We then exploit the learned attention patterns for Gaussian decoding to efficiently lift features. Extensive experiments on pose-free novel view synthesis, 3D open-vocabulary segmentation, and view-invariant feature aggregation demonstrate our approach's effectiveness. Results show that a compact yet geometrically meaningful representation is sufficient for high-quality scene reconstruction and understanding, achieving superior memory efficiency and feature fidelity compared to existing methods. --- ### Ultra-lightweight Neural Video Representation Compression Authors: Ho Man Kwan, Tianhao Peng, Ge Gao, Fan Zhang, Mike Nilsson, Andrew Gower, David Bull Published: 2025-12-03 Categories: cs.CV, eess.IV arXiv: https://arxiv.org/abs/2512.04019v1 Recent works have demonstrated the viability of utilizing over-fitted implicit neural representations (INRs) as alternatives to autoencoder-based models for neural video compression. Among these INR-based video codecs, Neural Video Representation Compression (NVRC) was the first to adopt a fully end-to-end compression framework that compresses INRs, achieving state-of-the-art performance. Moreover, some recently proposed lightweight INRs have shown comparable performance to their baseline codecs with computational complexity lower than 10kMACs/pixel. In this work, we extend NVRC toward lightweight representations, and propose NVRC-Lite, which incorporates two key changes. Firstly, we integrated multi-scale feature grids into our lightweight neural representation, and the use of higher resolution grids significantly improves the performance of INRs at low complexity. Secondly, we address the issue that existing INRs typically leverage autoregressive models for entropy coding: these are effective but impractical due to their slow coding speed. In this work, we propose an octree-based context model for entropy coding high-dimensional feature grids, which accelerates the entropy coding module of the model. Our experimental results demonstrate that NVRC-Lite outperforms C3, one of the best lightweight INR-based video codecs, with up to 21.03% and 23.06% BD-rate savings when measured in PSNR and MS-SSIM, respectively, while achieving 8.4x encoding and 2.5x decoding speedup. The implementation of NVRC-Lite will be made available. --- ### TARA Test-by-Adaptive-Ranks for Quantum Anomaly Detection with Conformal Prediction Guarantees Authors: Davut Emre Tasar, Ceren Ocal Tasar Published: 2025-12-03 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2512.04016v1 Quantum key distribution (QKD) security fundamentally relies on the ability to distinguish genuine quantum correlations from classical eavesdropper simulations, yet existing certification methods lack rigorous statistical guarantees under finite-sample conditions and adversarial scenarios. We introduce TARA (Test by Adaptive Ranks), a novel framework combining conformal prediction with sequential martingale testing for quantum anomaly detection that provides distribution-free validity guarantees. TARA offers two complementary approaches. TARA k, based on Kolmogorov Smirnov calibration against local hidden variable (LHV) null distributions, achieving ROC AUC = 0.96 for quantum-classical discrimination. And TARA-m, employing betting martingales for streaming detection with anytime valid type I error control that enables real time monitoring of quantum channels. We establish theoretical guarantees proving that under (context conditional) exchangeability, conformal p-values remain uniformly distributed even for strongly contextual quantum data, confirming that quantum contextuality does not break conformal prediction validity a result with implications beyond quantum certification to any application of distribution-free methods to nonclassical data. Extensive validation on both IBM Torino (superconducting, CHSH = 2.725) and IonQ Forte Enterprise (trapped ion, CHSH = 2.716) quantum processors demonstrates cross-platform robustness, achieving 36% security margins above the classical CHSH bound of 2. Critically, our framework reveals a methodological concern affecting quantum certification more broadly: same-distribution calibration can inflate detection performance by up to 44 percentage points compared to proper cross-distribution calibration, suggesting that prior quantum certification studies using standard train test splits may have systematically overestimated adversarial robustness. --- ### Learning Group Actions In Disentangled Latent Image Representations Authors: Farhana Hossain Swarnali, Miaomiao Zhang, Tonmoy Hossain Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.04015v2 Modeling group actions on latent representations enables controllable transformations of high-dimensional image data. Prior works applying group-theoretic priors or modeling transformations typically operate in the high-dimensional data space, where group actions apply uniformly across the entire input, making it difficult to disentangle the subspace that varies under transformations. While latent-space methods offer greater flexibility, they still require manual partitioning of latent variables into equivariant and invariant subspaces, limiting the ability to robustly learn and operate group actions within the representation space. To address this, we introduce a novel end-to-end framework that for the first time learns group actions on latent image manifolds, automatically discovering transformation-relevant structures without manual intervention. Our method uses learnable binary masks with straight-through estimation to dynamically partition latent representations into transformation-sensitive and invariant components. We formulate this within a unified optimization framework that jointly learns latent disentanglement and group transformation mappings. The framework can be seamlessly integrated with any standard encoder-decoder architecture. We validate our approach on five 2D/3D image datasets, demonstrating its ability to automatically learn disentangled latent factors for group actions in diverse data, while downstream classification tasks confirm the effectiveness of the learned representations. Our code is publicly available at https://github.com/farhanaswarnali/Learning-Group-Actions-In-Disentangled-Latent-Image-Representations . --- ### Learning Group Actions In Disentangled Latent Image Representations Authors: Farhana Hossain Swarnali, Miaomiao Zhang, Tonmoy Hossain Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.04015v1 Modeling group actions on latent representations enables controllable transformations of high-dimensional image data. Prior works applying group-theoretic priors or modeling transformations typically operate in the high-dimensional data space, where group actions apply uniformly across the entire input, making it difficult to disentangle the subspace that varies under transformations. While latent-space methods offer greater flexibility, they still require manual partitioning of latent variables into equivariant and invariant subspaces, limiting the ability to robustly learn and operate group actions within the representation space. To address this, we introduce a novel end-to-end framework that for the first time learns group actions on latent image manifolds, automatically discovering transformation-relevant structures without manual intervention. Our method uses learnable binary masks with straight-through estimation to dynamically partition latent representations into transformation-sensitive and invariant components. We formulate this within a unified optimization framework that jointly learns latent disentanglement and group transformation mappings. The framework can be seamlessly integrated with any standard encoder-decoder architecture. We validate our approach on five 2D/3D image datasets, demonstrating its ability to automatically learn disentangled latent factors for group actions in diverse data, while downstream classification tasks confirm the effectiveness of the learned representations. Our code is publicly available at https://github.com/farhanaswarnali/Learning-Group-Actions-In-Disentangled-Latent-Image-Representations . --- ### AugServe: Adaptive Request Scheduling for Augmented Large Language Model Inference Serving Authors: Ying Wang, Zhen Jin, Jiexiong Xu, Wenhai Lin, Yiquan Chen, Wenzhi Chen Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.04013v3 As augmented large language models (LLMs) with external tools become increasingly popular in web applications, improving augmented LLM inference serving efficiency and optimizing service-level objectives (SLOs) are critical for enhancing user experience. To achieve this, inference systems must maximize request handling within latency constraints, referred to as increasing effective throughput. However, existing systems face two major challenges: (i) reliance on first-come-first-served (FCFS) scheduling causes severe head-of-line blocking, leading to queuing delays exceeding the SLOs for many requests; and (ii) static batch token limit, which fails to adapt to fluctuating loads and hardware conditions. Both of these factors degrade effective throughput and service quality. This paper presents AugServe, an efficient inference framework designed to reduce queueing latency and enhance effective throughput for augmented LLM inference services. The core idea of AugServe is a two-stage adaptive request scheduling strategy. Specifically, AugServe combines the inference features of augmented LLM requests to optimize the order of scheduling decisions (stage I). These decisions are continuously refined with runtime information (stage II), adapting to both request characteristics and system capabilities. In addition, AugServe dynamically adjusts the token batching mechanism based on hardware status and real-time load, further enhancing throughput performance. Experimental results show that AugServe achieves 4.7x and 3.3x higher effective throughput than vLLM and InferCept, while reducing time-to-first-token (TTFT) by up to 96.3% and 95.0%, respectively. --- ### AugServe: Adaptive Request Scheduling for Augmented Large Language Model Inference Serving Authors: Ying Wang, Zhen Jin, Jiexiong Xu, Wenhai Lin, Yiquan Chen, Wenzhi Chen Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.04013v1 As augmented large language models (LLMs) with external tools become increasingly popular in web applications, improving augmented LLM inference serving efficiency and optimizing service-level objectives (SLOs) are critical for enhancing user experience. To achieve this, inference systems must maximize request handling within latency constraints, referred to as increasing effective throughput. However, existing systems face two major challenges: (i) reliance on first-come-first-served (FCFS) scheduling causes severe head-of-line blocking, leading to queuing delays exceeding the SLOs for many requests; and (ii) static batch token limit, which fails to adapt to fluctuating loads and hardware conditions. Both of these factors degrade effective throughput and service quality. This paper presents AugServe, an efficient inference framework designed to reduce queueing latency and enhance effective throughput for augmented LLM inference services. The core idea of AugServe is a two-stage adaptive request scheduling strategy. Specifically, AugServe combines the inference features of augmented LLM requests to optimize the order of scheduling decisions (stage I). These decisions are continuously refined with runtime information (stage II), adapting to both request characteristics and system capabilities. In addition, AugServe dynamically adjusts the token batching mechanism based on hardware status and real-time load, further enhancing throughput performance. Experimental results show that AugServe achieves 4.7-33.1x and 3.3-13.2x higher effective throughput than vLLM and InferCept, while reducing time-to-first-token (TTFT) by up to 96.3% and 95.0%, respectively. --- ### Emergent Outlier View Rejection in Visual Geometry Grounded Transformers Authors: Jisang Han, Sunghwan Hong, Jaewoo Jung, Wooseok Jang, Honggyu An, Qianqian Wang, Seungryong Kim, Chen Feng Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.04012v1 Reliable 3D reconstruction from in-the-wild image collections is often hindered by "noisy" images-irrelevant inputs with little or no view overlap with others. While traditional Structure-from-Motion pipelines handle such cases through geometric verification and outlier rejection, feed-forward 3D reconstruction models lack these explicit mechanisms, leading to degraded performance under in-the-wild conditions. In this paper, we discover that the existing feed-forward reconstruction model, e.g., VGGT, despite lacking explicit outlier-rejection mechanisms or noise-aware training, can inherently distinguish distractor images. Through an in-depth analysis under varying proportions of synthetic distractors, we identify a specific layer that naturally exhibits outlier-suppressing behavior. Further probing reveals that this layer encodes discriminative internal representations that enable an effective noise-filtering capability, which we simply leverage to perform outlier-view rejection in feed-forward 3D reconstruction without any additional fine-tuning or supervision. Extensive experiments on both controlled and in-the-wild datasets demonstrate that this implicit filtering mechanism is consistent and generalizes well across diverse scenarios. --- ### Efficient Public Verification of Private ML via Regularization Authors: Zoë Ruha Bell, Anvith Thudi, Olive Franzese-McLaughlin, Nicolas Papernot, Shafi Goldwasser Published: 2025-12-03 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2512.04008v2 Training with differential privacy (DP) guarantees dataset members that they cannot be identified by users of the released model. However, those data providers, and, in general, the public, lack methods to efficiently verify that models trained on their data satisfy DP guarantees. The amount of compute needed to verify DP guarantees for current algorithms scales with the amount of computation required to train the model. In this paper we design the first DP algorithm with near optimal privacy-utility trade-offs but whose DP guarantees can be verified cheaper than training. We focus on DP stochastic convex optimization (DP-SCO), where optimal privacy-utility trade-offs are known. Here we show we can obtain tight privacy-utility trade-offs by privately minimizing a series of regularized objectives and only using the standard DP composition bound. Crucially, this method can be verified with much less compute than training. This leads to the first known DP-SCO algorithm with near optimal privacy-utility whose DP verification scales better than training cost, significantly reducing verification costs on large datasets. --- ### Efficient Public Verification of Private ML via Regularization Authors: Zoë Ruha Bell, Anvith Thudi, Olive Franzese-McLaughlin, Nicolas Papernot, Shafi Goldwasser Published: 2025-12-03 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2512.04008v1 Training with differential privacy (DP) provides a guarantee to members in a dataset that they cannot be identified by users of the released model. However, those data providers, and, in general, the public, lack methods to efficiently verify that models trained on their data satisfy DP guarantees. The amount of compute needed to verify DP guarantees for current algorithms scales with the amount of compute required to train the model. In this paper we design the first DP algorithm with near optimal privacy-utility trade-offs but whose DP guarantees can be verified cheaper than training. We focus on DP stochastic convex optimization (DP-SCO), where optimal privacy-utility trade-offs are known. Here we show we can obtain tight privacy-utility trade-offs by privately minimizing a series of regularized objectives and only using the standard DP composition bound. Crucially, this method can be verified with much less compute than training. This leads to the first known DP-SCO algorithm with near optimal privacy-utility whose DP verification scales better than training cost, significantly reducing verification costs on large datasets. --- ### On the Temporality for Sketch Representation Learning Authors: Marcelo Isaias de Moraes Junior, Moacir Antonelli Ponti Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.04007v2 Sketches are simple human hand-drawn abstractions of complex scenes and real-world objects. Although the field of sketch representation learning has advanced significantly, there is still a gap in understanding the true relevance of the temporal aspect to the quality of these representations. This work investigates whether it is indeed justifiable to treat sketches as sequences, as well as which internal orders play a more relevant role. The results indicate that, although the use of traditional positional encodings is valid for modeling sketches as sequences, absolute coordinates consistently outperform relative ones. Furthermore, non-autoregressive decoders outperform their autoregressive counterparts. Finally, the importance of temporality was shown to depend on both the order considered and the task evaluated. --- ### On the Temporality for Sketch Representation Learning Authors: Marcelo Isaias de Moraes Junior, Moacir Antonelli Ponti Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.04007v1 Sketches are simple human hand-drawn abstractions of complex scenes and real-world objects. Although the field of sketch representation learning has advanced significantly, there is still a gap in understanding the true relevance of the temporal aspect to the quality of these representations. This work investigates whether it is indeed justifiable to treat sketches as sequences, as well as which internal orders play a more relevant role. The results indicate that, although the use of traditional positional encodings is valid for modeling sketches as sequences, absolute coordinates consistently outperform relative ones. Furthermore, non-autoregressive decoders outperform their autoregressive counterparts. Finally, the importance of temporality was shown to depend on both the order considered and the task evaluated. --- ### Diagonalizing the Softmax: Hadamard Initialization for Tractable Cross-Entropy Dynamics Authors: Connall Garrod, Jonathan P. Keating, Christos Thrampoulidis Published: 2025-12-03 Categories: cs.LG, math.OC, stat.ML arXiv: https://arxiv.org/abs/2512.04006v1 Cross-entropy (CE) training loss dominates deep learning practice, yet existing theory often relies on simplifications, either replacing it with squared loss or restricting to convex models, that miss essential behavior. CE and squared loss generate fundamentally different dynamics, and convex linear models cannot capture the complexities of non-convex optimization. We provide an in-depth characterization of multi-class CE optimization dynamics beyond the convex regime by analyzing a canonical two-layer linear neural network with standard-basis vectors as inputs: the simplest non-convex extension for which the implicit bias remained unknown. This model coincides with the unconstrained features model used to study neural collapse, making our work the first to prove that gradient flow on CE converges to the neural collapse geometry. We construct an explicit Lyapunov function that establishes global convergence, despite the presence of spurious critical points in the non-convex landscape. A key insight underlying our analysis is an inconspicuous finding: Hadamard Initialization diagonalizes the softmax operator, freezing the singular vectors of the weight matrices and reducing the dynamics entirely to their singular values. This technique opens a pathway for analyzing CE training dynamics well beyond our specific setting considered here. --- ### Physics-Embedded Gaussian Process for Traffic State Estimation Authors: Yanlin Chen, Kehua Chen, Yinhai Wang Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.04004v1 Traffic state estimation (TSE) becomes challenging when probe-vehicle penetration is low and observations are spatially sparse. Pure data-driven methods lack physical explanations and have poor generalization when observed data is sparse. In contrast, physical models have difficulty integrating uncertainties and capturing the real complexity of traffic. To bridge this gap, recent studies have explored combining them by embedding physical structure into Gaussian process. These approaches typically introduce the governing equations as soft constraints through pseudo-observations, enabling the integration of model structure within a variational framework. However, these methods rely heavily on penalty tuning and lack principled uncertainty calibration, which makes them sensitive to model mis-specification. In this work, we address these limitations by presenting a novel Physics-Embedded Gaussian Process (PEGP), designed to integrate domain knowledge with data-driven methods in traffic state estimation. Specifically, we design two multi-output kernels informed by classic traffic flow models, constructed via the explicit application of the linearized differential operator. Experiments on HighD, NGSIM show consistent improvements over non-physics baselines. PEGP-ARZ proves more reliable under sparse observation, while PEGP-LWR achieves lower errors with denser observation. Ablation study further reveals that PEGP-ARZ residuals align closely with physics and yield calibrated, interpretable uncertainty, whereas PEGP-LWR residuals are more orthogonal and produce nearly constant variance fields. This PEGP framework combines physical priors, uncertainty quantification, which can provide reliable support for TSE. --- ### Divide, then Ground: Adapting Frame Selection to Query Types for Long-Form Video Understanding Authors: Jialuo Li, Bin Li, Jiahao Li, Yan Lu Published: 2025-12-03 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.04000v2 The application of Large Multimodal Models (LMMs) to long-form video understanding is constrained by limited context lengths and the computationally prohibitive cost of processing dense video tokens. Consequently, recent research has focused on query-aware frame selection, methods that often incur significant computational overhead. This paper challenges the assumption that such complex search mechanisms are universally necessary. We first identify and validate a query typology distinguishing between global query and localized query. We demonstrate that while uniform sampling is both effective and efficient for global queries, localized queries indeed necessitate query-aware selection for optimal performance. Building on this insight, we propose DIG, a training-free frame selection framework that adapts its strategy based on the query type. Specifically,DIG employs efficient uniform sampling for global queries while activating a specialized pipeline to extract query-relevant frames for localized queries. Experiments on three long-form video understanding benchmarks demonstrate that DIG consistently outperforms existing baselines and robustly improves LMM performance, even when scaling the input frame count to 256. --- ### Divide, then Ground: Adapting Frame Selection to Query Types for Long-Form Video Understanding Authors: Jialuo Li, Bin Li, Jiahao Li, Yan Lu Published: 2025-12-03 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.04000v1 The application of Large Multimodal Models (LMMs) to long-form video understanding is constrained by limited context lengths and the computationally prohibitive cost of processing dense video tokens. Consequently, recent research has focused on query-aware frame selection, methods that often incur significant computational overhead. This paper challenges the assumption that such complex search mechanisms are universally necessary. We first identify and validate a query typology distinguishing between global query and localized query. We demonstrate that while uniform sampling is both effective and efficient for global queries, localized queries indeed necessitate query-aware selection for optimal performance. Building on this insight, we propose DIG, a training-free frame selection framework that adapts its strategy based on the query type. Specifically,DIG employs efficient uniform sampling for global queries while activating a specialized pipeline to extract query-relevant frames for localized queries. Experiments on three long-form video understanding benchmarks demonstrate that DIG consistently outperforms existing baselines and robustly improves LMM performance, even when scaling the input frame count to 256. --- ### Highly Efficient Test-Time Scaling for T2I Diffusion Models with Text Embedding Perturbation Authors: Hang Xu, Linjiang Huang, Feng Zhao Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03996v1 Test-time scaling (TTS) aims to achieve better results by increasing random sampling and evaluating samples based on rules and metrics. However, in text-to-image(T2I) diffusion models, most related works focus on search strategies and reward models, yet the impact of the stochastic characteristic of noise in T2I diffusion models on the method's performance remains unexplored. In this work, we analyze the effects of randomness in T2I diffusion models and explore a new format of randomness for TTS: text embedding perturbation, which couples with existing randomness like SDE-injected noise to enhance generative diversity and quality. We start with a frequency-domain analysis of these formats of randomness and their impact on generation, and find that these two randomness exhibit complementary behavior in the frequency domain: spatial noise favors low-frequency components (early steps), while text embedding perturbation enhances high-frequency details (later steps), thereby compensating for the potential limitations of spatial noise randomness in high-frequency manipulation. Concurrently, text embedding demonstrates varying levels of tolerance to perturbation across different dimensions of the generation process. Specifically, our method consists of two key designs: (1) Introducing step-based text embedding perturbation, combining frequency-guided noise schedules with spatial noise perturbation. (2) Adapting the perturbation intensity selectively based on their frequency-specific contributions to generation and tolerance to perturbation. Our approach can be seamlessly integrated into existing TTS methods and demonstrates significant improvements on multiple benchmarks with almost no additional computation. Code is available at \href{https://github.com/xuhang07/TEP-Diffusion}{https://github.com/xuhang07/TEP-Diffusion}. --- ### Artificial Microsaccade Compensation: Stable Vision for an Ornithopter Authors: Levi Burner, Guido de Croon, Yiannis Aloimonos Published: 2025-12-03 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2512.03995v1 Animals with foveated vision, including humans, experience microsaccades, small, rapid eye movements that they are not aware of. Inspired by this phenomenon, we develop a method for "Artificial Microsaccade Compensation". It can stabilize video captured by a tailless ornithopter that has resisted attempts to use camera-based sensing because it shakes at 12-20 Hz. Our approach minimizes changes in image intensity by optimizing over 3D rotation represented in SO(3). This results in a stabilized video, computed in real time, suitable for human viewing, and free from distortion. When adapted to hold a fixed viewing orientation, up to occasional saccades, it can dramatically reduce inter-frame motion while also benefiting from an efficient recursive update. When compared to Adobe Premier Pro's warp stabilizer, which is widely regarded as the best commercial video stabilization software available, our method achieves higher quality results while also running in real time. --- ### Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs Authors: Oren Rachmil, Avishag Shapira, Roy Betser, Omer Hofman, Itay Gershon, Asaf Shabtai, Yuval Elovici, Roman Vainshtein Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03994v4 As organizations increasingly deploy LLMs in sensitive domains such as legal, financial, and medical settings, ensuring alignment with internal organizational policies has become a priority. Existing content moderation frameworks remain largely confined to the safety domain and lack the robustness to capture nuanced organizational policies. LLM-as-a-judge and fine-tuning approaches, though flexible, introduce significant latency and training cost. To address these limitations, we frame policy violation detection as an out-of-distribution (OOD) problem in the model's activation space. We propose a training-free method that operates directly on the LLM internal representations, leveraging prior evidence that decision-relevant information is encoded within them. Inspired by whitening techniques, we derive policy-violation scores directly from normalized representations of LLM hidden activations. Our method requires only the policy text and a small number of illustrative samples, making it lightweight and easily deployable. We extensively evaluate our method across multiple LLMs and challenging policy benchmarks. It achieves up to 86.0% F1, outperforming both fine-tuned and LLM-as-a-judge baselines while requiring substantially less computation. Our code is publicly available at: https://github.com/FujitsuResearch/LLM-policy-violation-detection --- ### Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs Authors: Oren Rachmil, Roy Betser, Itay Gershon, Omer Hofman, Nitay Yakoby, Yuval Meron, Idan Yankelev, Asaf Shabtai, Yuval Elovici, Roman Vainshtein Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03994v1 Aligning proprietary large language models (LLMs) with internal organizational policies has become an urgent priority as organizations increasingly deploy LLMs in sensitive domains such as legal support, finance, and medical services. Beyond generic safety filters, enterprises require reliable mechanisms to detect policy violations within their regulatory and operational frameworks, where breaches can trigger legal and reputational risks. Existing content moderation frameworks, such as guardrails, remain largely confined to the safety domain and lack the robustness to capture nuanced organizational policies. LLM-as-a-judge and fine-tuning approaches, though flexible, introduce significant latency and lack interpretability. To address these limitations, we propose a training-free and efficient method that treats policy violation detection as an out-of-distribution (OOD) detection problem. Inspired by whitening techniques, we apply a linear transformation to decorrelate the model's hidden activations and standardize them to zero mean and unit variance, yielding near-identity covariance matrix. In this transformed space, we use the Euclidean norm as a compliance score to detect policy violations. The method requires only the policy text and a small number of illustrative samples, which makes it light-weight and easily deployable. On a challenging policy benchmark, our approach achieves state-of-the-art results, surpassing both existing guardrails and fine-tuned reasoning models. This work provides organizations with a practical and statistically grounded framework for policy-aware oversight of LLMs, advancing the broader goal of deployable AI governance. Code is available at: https://tinyurl.com/policy-violation-detection --- ### Value-Guided Iterative Refinement and the DIQ-H Benchmark for Evaluating VLM Robustness Authors: Hanwen Wan, Zexin Lin, Yixuan Deng, Xiaoqiang Ji Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03992v2 Vision-Language Models (VLMs) are essential for embodied AI and safety-critical applications, such as robotics and autonomous systems. However, existing benchmarks primarily focus on static or curated visual inputs, neglecting the challenges posed by adversarial conditions, value misalignment, and error propagation in continuous deployment. Current benchmarks either overlook the impact of real-world perturbations, or fail to account for the cumulative effect of inconsistent reasoning over time. To address these gaps, we introduce the Degraded Image Quality Leading to Hallucinations (DIQ-H) benchmark, the first to evaluate VLMs under adversarial visual conditions in continuous sequences. DIQ-H simulates real-world stressors including motion blur, sensor noise, and compression artifacts, and measures how these corruptions lead to persistent errors and misaligned outputs across time. The benchmark explicitly models error propagation and its long-term value consistency. To enhance scalability and reduce costs for safety-critical evaluation, we propose the Value-Guided Iterative Refinement (VIR) framework, which automates the generation of high-quality, ethically aligned ground truth annotations. VGIR leverages lightweight VLMs to detect and refine value misalignment, improving accuracy from 72.2% to 83.3%, representing a 15.3% relative improvement. The DIQ-H benchmark and VGIR framework provide a robust platform for embodied AI safety assessment, revealing vulnerabilities in error recovery, ethical consistency, and temporal value alignment. --- ### DIQ-H: Evaluating Hallucination Persistence in VLMs Under Temporal Visual Degradation Authors: Zexin Lin, Hawen Wan, Yebin Zhong, Xiaoqiang Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03992v1 Vision-Language Models (VLMs) deployed in safety-critical applications such as autonomous driving must handle continuous visual streams under imperfect conditions. However, existing benchmarks focus on static, high-quality images and ignore temporal degradation and error propagation, which are critical failure modes where transient visual corruption induces hallucinations that persist across subsequent frames. We introduce DIQ-H, the first benchmark for evaluating VLM robustness under dynamic visual degradation in temporal sequences. DIQ-H applies physics-based corruptions including motion blur, sensor noise, and compression artifacts, and measures hallucination persistence, error recovery, and temporal consistency through multi-turn question-answering tasks. To enable scalable annotation, we propose Uncertainty-Guided Iterative Refinement (UIR), which generates reliable pseudo-ground-truth using lightweight VLMs with uncertainty filtering, achieving a 15.3 percent accuracy improvement. Experiments on 16 state-of-the-art VLMs reveal substantial robustness gaps: even advanced models such as GPT-4o achieve only a 78.5 percent recovery rate, while open-source models struggle with temporal consistency at less than 60 percent. DIQ-H provides a comprehensive platform for evaluating VLM reliability in real-world deployments. --- ### Teaching Old Tokenizers New Words: Efficient Tokenizer Adaptation for Pre-trained Models Authors: Taido Purason, Pavel Chizhov, Ivan P. Yamshchikov, Mark Fishel Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03989v2 Tokenizer adaptation plays an important role in adapting pre-trained language models to new domains or languages. In this work, we address two complementary aspects of this process: vocabulary extension and pruning. The common approach to extension trains a new tokenizer on domain-specific text and appends the tokens that do not overlap with the existing vocabulary, which often results in many tokens that are unreachable or never used. We propose continued BPE training that extends a pre-trained tokenizer by continuing the BPE merge learning process on new data. Experiments across multiple languages and model families show that this approach improves tokenization efficiency and leads to better utilization of added vocabulary. We also introduce leaf-based vocabulary pruning, which removes redundant tokens while preserving model quality. Together, these methods provide practical tools for controlled vocabulary modification, which we release as an open-source toolkit. --- ### Teaching Old Tokenizers New Words: Efficient Tokenizer Adaptation for Pre-trained Models Authors: Taido Purason, Pavel Chizhov, Ivan P. Yamshchikov, Mark Fishel Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03989v1 Tokenizer adaptation plays an important role in transferring pre-trained language models to new domains or languages. In this work, we address two complementary aspects of this process: vocabulary extension and pruning. The common approach to extension trains a new tokenizer on domain-specific text and appends the tokens that do not overlap with the existing vocabulary, which often results in many tokens that are unreachable or never used. We propose continued BPE training, which adapts a pre-trained tokenizer by continuing the BPE merge learning process on new data. Experiments across multiple languages and model families show that this approach improves tokenization efficiency and leads to better utilization of added vocabulary. We also introduce leaf-based vocabulary pruning, which removes redundant tokens while preserving model quality. Together, these methods provide practical tools for controlled vocabulary modification, which we release as an open-source package. --- ### DirectDrag: High-Fidelity, Mask-Free, Prompt-Free Drag-based Image Editing via Readout-Guided Feature Alignment Authors: Sheng-Hao Liao, Shang-Fu Chen, Tai-Ming Huang, Wen-Huang Cheng, Kai-Lung Hua Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03981v1 Drag-based image editing using generative models provides intuitive control over image structures. However, existing methods rely heavily on manually provided masks and textual prompts to preserve semantic fidelity and motion precision. Removing these constraints creates a fundamental trade-off: visual artifacts without masks and poor spatial control without prompts. To address these limitations, we propose DirectDrag, a novel mask- and prompt-free editing framework. DirectDrag enables precise and efficient manipulation with minimal user input while maintaining high image fidelity and accurate point alignment. DirectDrag introduces two key innovations. First, we design an Auto Soft Mask Generation module that intelligently infers editable regions from point displacement, automatically localizing deformation along movement paths while preserving contextual integrity through the generative model's inherent capacity. Second, we develop a Readout-Guided Feature Alignment mechanism that leverages intermediate diffusion activations to maintain structural consistency during point-based edits, substantially improving visual fidelity. Despite operating without manual mask or prompt, DirectDrag achieves superior image quality compared to existing methods while maintaining competitive drag accuracy. Extensive experiments on DragBench and real-world scenarios demonstrate the effectiveness and practicality of DirectDrag for high-quality, interactive image manipulation. Project Page: https://frakw.github.io/DirectDrag/. Code is available at: https://github.com/frakw/DirectDrag. --- ### BlurDM: A Blur Diffusion Model for Image Deblurring Authors: Jin-Ting He, Fu-Jen Tsai, Yan-Tsung Peng, Min-Hung Chen, Chia-Wen Lin, Yen-Yu Lin Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03979v3 Diffusion models show promise for dynamic scene deblurring; however, existing studies often fail to leverage the intrinsic nature of the blurring process within diffusion models, limiting their full potential. To address it, we present a Blur Diffusion Model (BlurDM), which seamlessly integrates the blur formation process into diffusion for image deblurring. Observing that motion blur stems from continuous exposure, BlurDM implicitly models the blur formation process through a dual-diffusion forward scheme, diffusing both noise and blur onto a sharp image. During the reverse generation process, we derive a dual denoising and deblurring formulation, enabling BlurDM to recover the sharp image by simultaneously denoising and deblurring, given pure Gaussian noise conditioned on the blurred image as input. Additionally, to efficiently integrate BlurDM into deblurring networks, we perform BlurDM in the latent space, forming a flexible prior generation network for deblurring. Extensive experiments demonstrate that BlurDM significantly and consistently enhances existing deblurring methods on four benchmark datasets. The project page is available at https://jin-ting-he.github.io/BlurDM/. --- ### BlurDM: A Blur Diffusion Model for Image Deblurring Authors: Jin-Ting He, Fu-Jen Tsai, Yan-Tsung Peng, Min-Hung Chen, Chia-Wen Lin, Yen-Yu Lin Published: 2025-12-03 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.03979v1 Diffusion models show promise for dynamic scene deblurring; however, existing studies often fail to leverage the intrinsic nature of the blurring process within diffusion models, limiting their full potential. To address it, we present a Blur Diffusion Model (BlurDM), which seamlessly integrates the blur formation process into diffusion for image deblurring. Observing that motion blur stems from continuous exposure, BlurDM implicitly models the blur formation process through a dual-diffusion forward scheme, diffusing both noise and blur onto a sharp image. During the reverse generation process, we derive a dual denoising and deblurring formulation, enabling BlurDM to recover the sharp image by simultaneously denoising and deblurring, given pure Gaussian noise conditioned on the blurred image as input. Additionally, to efficiently integrate BlurDM into deblurring networks, we perform BlurDM in the latent space, forming a flexible prior generation network for deblurring. Extensive experiments demonstrate that BlurDM significantly and consistently enhances existing deblurring methods on four benchmark datasets. The source code is available at https://github.com/Jin-Ting-He/BlurDM. --- ### Adapting Large Language Models to Low-Resource Tibetan: A Two-Stage Continual and Supervised Fine-Tuning Study Authors: Lifeng Chen, Ryan Lai, Tianming Liu Published: 2025-12-03 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.03976v1 Adapting large language models (LLMs) to low-resource languages remains a major challenge due to data scarcity and cross-lingual drift. This work presents a two-stage adaptation of Qwen2.5-3B to Tibetan, a morphologically rich and underrepresented language. We employ Continual Pretraining (CPT) to establish Tibetan linguistic grounding, followed by Supervised Fine-Tuning (SFT) for task and translation specialization. Empirical evaluations demonstrate a consistent decrease in perplexity (from 2.98 $\rightarrow$ 1.54) and substantial improvements in Chinese$\rightarrow$Tibetan translation quality (BLEU: 0.046 $\rightarrow$ 0.261; chrF: 2.2 $\rightarrow$ 6.6). Layer-wise analysis across 435 layers in Qwen3-4B reveals that adaptation primarily concentrates on embedding and output heads, with mid--late MLP projections encoding domain-specific transformations. Our findings suggest that CPT constructs a Tibetan semantic manifold while SFT sharpens task alignment with minimal representational disruption. This study provides the first quantitative exploration of Tibetan adaptation dynamics for LLMs, and offers an open, reproducible framework for extending multilingual foundation models to low-resource settings. --- ### Sponsored Questions and How to Auction Them Authors: Kshipra Bhawalkar, Alexandros Psomas, Di Wang Published: 2025-12-03 Categories: cs.GT, cs.AI arXiv: https://arxiv.org/abs/2512.03975v1 Online platforms connect users with relevant products and services using ads. A key challenge is that a user's search query often leaves their true intent ambiguous. Typically, platforms passively predict relevance based on available signals and in some cases offer query refinements. The shift from traditional search to conversational AI provides a new approach. When a user's query is ambiguous, a Large Language Model (LLM) can proactively offer several clarifying follow-up prompts. In this paper we consider the following: what if some of these follow-up prompts can be ``sponsored,'' i.e., selected for their advertising potential. How should these ``suggestion slots'' be allocated? And, how does this new mechanism interact with the traditional ad auction that might follow? This paper introduces a formal model for designing and analyzing these interactive platforms. We use this model to investigate a critical engineering choice: whether it is better to build an end-to-end pipeline that jointly optimizes the user interaction and the final ad auction, or to decouple them into separate mechanisms for the suggestion slots and another for the subsequent ad slot. We show that the VCG mechanism can be adopted to jointly optimize the sponsored suggestion and the ads that follow; while this mechanism is more complex, it achieves outcomes that are efficient and truthful. On the other hand, we prove that the simple-to-implement modular approach suffers from strategic inefficiency: its Price of Anarchy is unbounded. --- ### Refining Machine Learning Potentials through Thermodynamic Theory of Phase Transitions Authors: Paul Fuchs, Julija Zavadlav Published: 2025-12-03 Categories: physics.comp-ph, cs.LG arXiv: https://arxiv.org/abs/2512.03974v1 Foundational Machine Learning Potentials can resolve the accuracy and transferability limitations of classical force fields. They enable microscopic insights into material behavior through Molecular Dynamics simulations, which can crucially expedite material design and discovery. However, insufficiently broad and systematically biased reference data affect the predictive quality of the learned models. Often, these models exhibit significant deviations from experimentally observed phase transition temperatures, in the order of several hundred kelvins. Thus, fine-tuning is necessary to achieve adequate accuracy in many practical problems. This work proposes a fine-tuning strategy via top-down learning, directly correcting the wrongly predicted transition temperatures to match the experimental reference data. Our approach leverages the Differentiable Trajectory Reweighting algorithm to minimize the free energy differences between phases at the experimental target pressures and temperatures. We demonstrate that our approach can accurately correct the phase diagram of pure Titanium in a pressure range of up to 5 GPa, matching the experimental reference within tenths of kelvins and improving the liquid-state diffusion constant. Our approach is model-agnostic, applicable to multi-component systems with solid-solid and solid-liquid transitions, and compliant with top-down training on other experimental properties. Therefore, our approach can serve as an essential step towards highly accurate application-specific and foundational machine learning potentials. --- ### Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning Authors: Franki Nguimatsia Tiofack, Théotime Le Hellard, Fabian Schramm, Nicolas Perrin-Gilbert, Justin Carpentier Published: 2025-12-03 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.03973v2 Offline reinforcement learning often relies on behavior regularization that enforces policies to remain close to the dataset distribution. However, such approaches fail to distinguish between high-value and low-value actions in their regularization components. We introduce Guided Flow Policy (GFP), which couples a multi-step flow-matching policy with a distilled one-step actor. The actor directs the flow policy through weighted behavior cloning to focus on cloning high-value actions from the dataset rather than indiscriminately imitating all state-action pairs. In turn, the flow policy constrains the actor to remain aligned with the dataset's best transitions while maximizing the critic. This mutual guidance enables GFP to achieve state-of-the-art performance across 144 state and pixel-based tasks from the OGBench, Minari, and D4RL benchmarks, with substantial gains on suboptimal datasets and challenging tasks. Webpage: https://simple-robotics.github.io/publications/guided-flow-policy/ --- ### Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning Authors: Franki Nguimatsia Tiofack, Théotime Le Hellard, Fabian Schramm, Nicolas Perrin-Gilbert, Justin Carpentier Published: 2025-12-03 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.03973v1 Offline reinforcement learning often relies on behavior regularization that enforces policies to remain close to the dataset distribution. However, such approaches fail to distinguish between high-value and low-value actions in their regularization components. We introduce Guided Flow Policy (GFP), which couples a multi-step flow-matching policy with a distilled one-step actor. The actor directs the flow policy through weighted behavior cloning to focus on cloning high-value actions from the dataset rather than indiscriminately imitating all state-action pairs. In turn, the flow policy constrains the actor to remain aligned with the dataset's best transitions while maximizing the critic. This mutual guidance enables GFP to achieve state-of-the-art performance across 144 state and pixel-based tasks from the OGBench, Minari, and D4RL benchmarks, with substantial gains on suboptimal datasets and challenging tasks. Webpage: https://simple-robotics.github.io/publications/guided-flow-policy/ --- ### Technical Report on Text Dataset Distillation Authors: Keith Ando Ogawa, Bruno Lopes Yamamoto, Lucas Lauton de Alcantara, Victor Zacarias, Edson Bollis, Lucas Pellicer, Rosimeire Pereira Costa, Anna Helena Reali Costa, Artur Jordao Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03967v1 In the vision domain, dataset distillation arises as a technique to condense a large dataset into a smaller synthetic one that exhibits a similar result in the training process. While image data presents an extensive literature of distillation methods, text dataset distillation has fewer works in comparison. Text dataset distillation initially grew as an adaptation of efforts from the vision universe, as the particularities of the modality became clear obstacles, it rose into a separate branch of research. Several milestones mark the development of this area, such as the introduction of methods that use transformer models, the generation of discrete synthetic text, and the scaling to decoder-only models with over 1B parameters. Despite major advances in modern approaches, the field remains in a maturing phase, with room for improvement on benchmarking standardization, approaches to overcome the discrete nature of text, handling complex tasks, and providing explicit examples of real-world applications. In this report, we review past and recent advances in dataset distillation for text, highlighting different distillation strategies, key contributions, and general challenges. --- ### Training for Identity, Inference for Controllability: A Unified Approach to Tuning-Free Face Personalization Authors: Lianyu Pang, Ji Zhou, Qiping Wang, Baoquan Zhao, Zhenguo Yang, Qing Li, Xudong Mao Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03964v1 Tuning-free face personalization methods have developed along two distinct paradigms: text embedding approaches that map facial features into the text embedding space, and adapter-based methods that inject features through auxiliary cross-attention layers. While both paradigms have shown promise, existing methods struggle to simultaneously achieve high identity fidelity and flexible text controllability. We introduce UniID, a unified tuning-free framework that synergistically integrates both paradigms. Our key insight is that when merging these approaches, they should mutually reinforce only identity-relevant information while preserving the original diffusion prior for non-identity attributes. We realize this through a principled training-inference strategy: during training, we employ an identity-focused learning scheme that guides both branches to capture identity features exclusively; at inference, we introduce a normalized rescaling mechanism that recovers the text controllability of the base diffusion model while enabling complementary identity signals to enhance each other. This principled design enables UniID to achieve high-fidelity face personalization with flexible text controllability. Extensive experiments against six state-of-the-art methods demonstrate that UniID achieves superior performance in both identity preservation and text controllability. Code will be available at https://github.com/lyuPang/UniID --- ### TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning Authors: Tao Wu, Li Yang, Gen Zhan, Yabin Zhang, Yiting Liao, Junlin Li, Deliang Fu, Li Zhang, Limin Wang Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03963v3 Enhancing the temporal understanding of Multimodal Large Language Models (MLLMs) is essential for advancing long-form video analysis, enabling tasks such as temporal localization, action detection, and time-sensitive question answering. While reinforcement learning (RL) has recently been explored for improving temporal reasoning, existing approaches are often confined to limited task types and data, restricting their generalization across diverse temporal understanding scenarios. To address this challenge, we present TempR1, a temporal-aware multi-task reinforcement learning framework that systematically strengthens MLLMs' temporal comprehension. We curate a multi-task corpus that exposes the model to diverse temporal structures and semantics, and build upon the Group Relative Policy Optimization (GRPO) algorithm to achieve stable and effective cross-task optimization. Specifically, we categorize temporal tasks into three correspondence types between predicted intervals and ground-truth instances, and design tailored localization rewards for each, enabling TempR1 to capture fine-grained temporal dependencies and adapt to different temporal patterns. Extensive experiments demonstrate that TempR1 attains state-of-the-art performance across multiple benchmarks. Moreover, its joint optimization over complementary tasks yields a strong synergistic effect, enhancing both generalization and single-task performance, establishing a scalable and principled paradigm for temporal reasoning in MLLMs. --- ### TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning Authors: Tao Wu, Li Yang, Gen Zhan, Yiting Liao, Junlin Li, Deliang Fu, Li Zhang, Limin Wang Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03963v1 Enhancing the temporal understanding of Multimodal Large Language Models (MLLMs) is essential for advancing long-form video analysis, enabling tasks such as temporal localization, action detection, and time-sensitive question answering. While reinforcement learning (RL) has recently been explored for improving temporal reasoning, existing approaches are often confined to limited task types and data, restricting their generalization across diverse temporal understanding scenarios. To address this challenge, we present TempR1, a temporal-aware multi-task reinforcement learning framework that systematically strengthens MLLMs' temporal comprehension. We curate a multi-task corpus that exposes the model to diverse temporal structures and semantics, and build upon the Group Relative Policy Optimization (GRPO) algorithm to achieve stable and effective cross-task optimization. Specifically, we categorize temporal tasks into three correspondence types between predicted intervals and ground-truth instances, and design tailored localization rewards for each, enabling TempR1 to capture fine-grained temporal dependencies and adapt to different temporal patterns. Extensive experiments demonstrate that TempR1 attains state-of-the-art performance across multiple benchmarks. Moreover, its joint optimization over complementary tasks yields a strong synergistic effect, enhancing both generalization and single-task performance, establishing a scalable and principled paradigm for temporal reasoning in MLLMs. --- ### Tada-DIP: Input-adaptive Deep Image Prior for One-shot 3D Image Reconstruction Authors: Evan Bell, Shijun Liang, Ismail Alkhouri, Saiprasad Ravishankar Published: 2025-12-03 Categories: eess.IV, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.03962v1 Deep Image Prior (DIP) has recently emerged as a promising one-shot neural-network based image reconstruction method. However, DIP has seen limited application to 3D image reconstruction problems. In this work, we introduce Tada-DIP, a highly effective and fully 3D DIP method for solving 3D inverse problems. By combining input-adaptation and denoising regularization, Tada-DIP produces high-quality 3D reconstructions while avoiding the overfitting phenomenon that is common in DIP. Experiments on sparse-view X-ray computed tomography reconstruction validate the effectiveness of the proposed method, demonstrating that Tada-DIP produces much better reconstructions than training-data-free baselines and achieves reconstruction performance on par with a supervised network trained using a large dataset with fully-sampled volumes. --- ### Benchmark for Planning and Control with Large Language Model Agents: Blocksworld with Model Context Protocol Authors: Niklas Jobs, Luis Miguel Vieira da Silva, Jayanth Somashekaraiah, Maximilian Weigand, David Kube, Felix Gehlhoff Published: 2025-12-03 Categories: cs.AI, cs.ET arXiv: https://arxiv.org/abs/2512.03955v1 Industrial automation increasingly requires flexible control strategies that can adapt to changing tasks and environments. Agents based on Large Language Models (LLMs) offer potential for such adaptive planning and execution but lack standardized benchmarks for systematic comparison. We introduce a benchmark with an executable simulation environment representing the Blocksworld problem providing five complexity categories. By integrating the Model Context Protocol (MCP) as a standardized tool interface, diverse agent architectures can be connected to and evaluated against the benchmark without implementation-specific modifications. A single-agent implementation demonstrates the benchmark's applicability, establishing quantitative metrics for comparison of LLM-based planning and execution approaches. --- ### Is Lying Only Sinful in Islam? Exploring Religious Bias in Multilingual Large Language Models Across Major Religions Authors: Kazi Abrab Hossain, Jannatul Somiya Mahmud, Maria Hossain Tuli, Anik Mitra, S. M. Taiabul Haque, Farig Y. Sadeque Published: 2025-12-03 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2512.03943v1 While recent developments in large language models have improved bias detection and classification, sensitive subjects like religion still present challenges because even minor errors can result in severe misunderstandings. In particular, multilingual models often misrepresent religions and have difficulties being accurate in religious contexts. To address this, we introduce BRAND: Bilingual Religious Accountable Norm Dataset, which focuses on the four main religions of South Asia: Buddhism, Christianity, Hinduism, and Islam, containing over 2,400 entries, and we used three different types of prompts in both English and Bengali. Our results indicate that models perform better in English than in Bengali and consistently display bias toward Islam, even when answering religion-neutral questions. These findings highlight persistent bias in multilingual models when similar questions are asked in different languages. We further connect our findings to the broader issues in HCI regarding religion and spirituality. --- ### MUT3R: Motion-aware Updating Transformer for Dynamic 3D Reconstruction Authors: Guole Shen, Tianchen Deng, Xingrui Qin, Nailin Wang, Jianyu Wang, Yanbo Wang, Yongtao Chen, Hesheng Wang, Jingchuan Wang Published: 2025-12-03 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2512.03939v1 Recent stateful recurrent neural networks have achieved remarkable progress on static 3D reconstruction but remain vulnerable to motion-induced artifacts, where non-rigid regions corrupt attention propagation between the spatial memory and image feature. By analyzing the internal behaviors of the state and image token updating mechanism, we find that aggregating self-attention maps across layers reveals a consistent pattern: dynamic regions are naturally down-weighted, exposing an implicit motion cue that the pretrained transformer already encodes but never explicitly uses. Motivated by this observation, we introduce MUT3R, a training-free framework that applies the attention-derived motion cue to suppress dynamic content in the early layers of the transformer during inference. Our attention-level gating module suppresses the influence of dynamic regions before their artifacts propagate through the feature hierarchy. Notably, we do not retrain or fine-tune the model; we let the pretrained transformer diagnose its own motion cues and correct itself. This early regulation stabilizes geometric reasoning in streaming scenarios and leads to improvements in temporal consistency and camera pose robustness across multiple dynamic benchmarks, offering a simple and training-free pathway toward motion-aware streaming reconstruction. --- ### Beyond the Ground Truth: Enhanced Supervision for Image Restoration Authors: Donghun Ryou, Inju Ha, Sanghyeok Chu, Bohyung Han Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03932v3 Deep learning-based image restoration has achieved significant success. However, when addressing real-world degradations, model performance is limited by the quality of groundtruth images in datasets due to practical constraints in data acquisition. To address this limitation, we propose a novel framework that enhances existing ground truth images to provide higher-quality supervision for real-world restoration. Our framework generates perceptually enhanced ground truth images using super-resolution by incorporating adaptive frequency masks, which are learned by a conditional frequency mask generator. These masks guide the optimal fusion of frequency components from the original ground truth and its super-resolved variants, yielding enhanced ground truth images. This frequency-domain mixup preserves the semantic consistency of the original content while selectively enriching perceptual details, preventing hallucinated artifacts that could compromise fidelity. The enhanced ground truth images are used to train a lightweight output refinement network that can be seamlessly integrated with existing restoration models. Extensive experiments demonstrate that our approach improves the quality of restored images. We further validate the effectiveness of both supervision enhancement and output refinement through user studies. --- ### Beyond the Ground Truth: Enhanced Supervision for Image Restoration Authors: Donghun Ryou, Inju Ha, Sanghyeok Chu, Bohyung Han Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03932v1 Deep learning-based image restoration has achieved significant success. However, when addressing real-world degradations, model performance is limited by the quality of ground-truth images in datasets due to practical constraints in data acquisition. To address this limitation, we propose a novel framework that enhances existing ground truth images to provide higher-quality supervision for real-world restoration. Our framework generates perceptually enhanced ground truth images using super-resolution by incorporating adaptive frequency masks, which are learned by a conditional frequency mask generator. These masks guide the optimal fusion of frequency components from the original ground truth and its super-resolved variants, yielding enhanced ground truth images. This frequency-domain mixup preserves the semantic consistency of the original content while selectively enriching perceptual details, preventing hallucinated artifacts that could compromise fidelity. The enhanced ground truth images are used to train a lightweight output refinement network that can be seamlessly integrated with existing restoration models. Extensive experiments demonstrate that our approach consistently improves the quality of restored images. We further validate the effectiveness of both supervision enhancement and output refinement through user studies. Code is available at https://github.com/dhryougit/Beyond-the-Ground-Truth. --- ### Autonomous Agents and Policy Compliance: A Framework for Reasoning About Penalties Authors: Vineel Tummala, Daniela Inclezan Published: 2025-12-03 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.03931v1 This paper presents a logic programming-based framework for policy-aware autonomous agents that can reason about potential penalties for non-compliance and act accordingly. While prior work has primarily focused on ensuring compliance, our approach considers scenarios where deviating from policies may be necessary to achieve high-stakes goals. Additionally, modeling non-compliant behavior can assist policymakers by simulating realistic human decision-making. Our framework extends Gelfond and Lobo's Authorization and Obligation Policy Language (AOPL) to incorporate penalties and integrates Answer Set Programming (ASP) for reasoning. Compared to previous approaches, our method ensures well-formed policies, accounts for policy priorities, and enhances explainability by explicitly identifying rule violations and their consequences. Building on the work of Harders and Inclezan, we introduce penalty-based reasoning to distinguish between non-compliant plans, prioritizing those with minimal repercussions. To support this, we develop an automated translation from the extended AOPL into ASP and refine ASP-based planning algorithms to account for incurred penalties. Experiments in two domains demonstrate that our framework generates higher-quality plans that avoid harmful actions while, in some cases, also improving computational efficiency. These findings underscore its potential for enhancing autonomous decision-making and informing policy refinement. Under consideration in Theory and Practice of Logic Programming (TPLP). --- ### Density-Informed VAE (DiVAE): Reliable Log-Prior Probability via Density Alignment Regularization Authors: Michele Alessi, Alessio Ansuini, Alex Rodriguez Published: 2025-12-03 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.03928v1 We introduce Density-Informed VAE (DiVAE), a lightweight, data-driven regularizer that aligns the VAE log-prior probability $\log p_Z(z)$ with a log-density estimated from data. Standard VAEs match latents to a simple prior, overlooking density structure in the data-space. DiVAE encourages the encoder to allocate posterior mass in proportion to data-space density and, when the prior is learnable, nudges the prior toward high-density regions. This is realized by adding a robust, precision-weighted penalty to the ELBO, incurring negligible computational overhead. On synthetic datasets, DiVAE (i) improves distributional alignment of latent log-densities to its ground truth counterpart, (ii) improves prior coverage, and (iii) yields better OOD uncertainty calibration. On MNIST, DiVAE improves alignment of the prior with external estimates of the density, providing better interpretability, and improves OOD detection for learnable priors. --- ### Quantum-Classical Physics-Informed Neural Networks for Solving Reservoir Seepage Equations Authors: Xiang Rao, Yina Liu, Yuxuan Shen Published: 2025-12-03 Categories: cs.LG, math.NA, physics.comp-ph arXiv: https://arxiv.org/abs/2512.03923v3 Solving reservoir seepage partial differential equations (PDEs) is critical for hydrocarbon development optimization and production forecasting. Conventional numerical solvers suffer from mesh errors and high computational cost. Standard Physics-Informed Neural Networks (PINNs) are limited by inefficient parameters. They also face weak high-dimensional representation and poor nonlinear fitting ability. We adapt the Discrete Variable (DV)-Circuit Quantum-Classical Physics-Informed Neural Network (QCPINN) to tackle these drawbacks. This work applies the method to four representative reservoir seepage models. The models cover heterogeneous single-phase pressure diffusion. They also include nonlinear Buckley-Leverett two-phase waterflooding. Another case is adsorption-coupled convection-diffusion compositional flow. The last one refers to fully coupled pressure-saturation two-phase flow with exponential permeability heterogeneity. The QCPINN combines classical preprocessing and postprocessing networks with a DV quantum core. It leverages quantum entanglement and superposition to boost high-dimensional feature mapping. Physical constraints are embedded within the framework to guarantee physically consistent solutions. We evaluate three quantum circuit topologies including Cascade, Cross-mesh and Alternate. Numerical tests prove QCPINNs outperform classical PINNs in prediction accuracy, and the circuit topologies show case-specific performance with clear applicable ranges. This work verifies QCPINN's practicability for reservoir engineering. It bridges quantum computing research with oilfield industrial applications. --- ### Quantum-Classical Physics-Informed Neural Networks for Solving Reservoir Seepage Equations Authors: Xiang Rao, Yina Liu, Yuxuan Shen Published: 2025-12-03 Categories: cs.LG, math.NA, physics.comp-ph arXiv: https://arxiv.org/abs/2512.03923v1 Solving partial differential equations (PDEs) for reservoir seepage is critical for optimizing oil and gas field development and predicting production performance. Traditional numerical methods suffer from mesh-dependent errors and high computational costs, while classical Physics-Informed Neural Networks (PINNs) face bottlenecks in parameter efficiency, high-dimensional expression, and strong nonlinear fitting. To address these limitations, we propose a Discrete Variable (DV)-Circuit Quantum-Classical Physics-Informed Neural Network (QCPINN) and apply it to three typical reservoir seepage models for the first time: the pressure diffusion equation for heterogeneous single-phase flow, the nonlinear Buckley-Leverett (BL) equation for two-phase waterflooding, and the convection-diffusion equation for compositional flow considering adsorption. The QCPINN integrates classical preprocessing/postprocessing networks with a DV quantum core, leveraging quantum superposition and entanglement to enhance high-dimensional feature mapping while embedding physical constraints to ensure solution consistency. We test three quantum circuit topologies (Cascade, Cross-mesh, Alternate) and demonstrate through numerical experiments that QCPINNs achieve high prediction accuracy with fewer parameters than classical PINNs. Specifically, the Alternate topology outperforms others in heterogeneous single-phase flow and two-phase BL equation simulations, while the Cascade topology excels in compositional flow with convection-dispersion-adsorption coupling. Our work verifies the feasibility of QCPINN for reservoir engineering applications, bridging the gap between quantum computing research and industrial practice in oil and gas engineering. --- ### UniMo: Unifying 2D Video and 3D Human Motion with an Autoregressive Framework Authors: Youxin Pang, Yong Zhang, Ruizhi Shao, Xiang Deng, Feng Gao, Xu Xiaoming, Xiaoming Wei, Yebin Liu Published: 2025-12-03 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.03918v1 We propose UniMo, an innovative autoregressive model for joint modeling of 2D human videos and 3D human motions within a unified framework, enabling simultaneous generation and understanding of these two modalities for the first time. Current methods predominantly focus on generating one modality given another as the condition or integrating either of them with other modalities such as text and audio. Unifying 2D videos and 3D motions for simultaneous optimization and generation remains largely unexplored, presenting significant challenges due to their substantial structural and distributional differences. Inspired by the LLM's ability to unify different modalities, our method models videos and 3D motions as a unified tokens sequence, utilizing separate embedding layers to mitigate distribution gaps. Additionally, we devise a sequence modeling strategy that integrates two distinct tasks within a single framework, proving the effectiveness of unified modeling. Moreover, to efficiently align with visual tokens and preserve 3D spatial information, we design a novel 3D motion tokenizer with a temporal expansion strategy, using a single VQ-VAE to produce quantized motion tokens. It features multiple expert decoders that handle body shapes, translation, global orientation, and body poses for reliable 3D motion reconstruction. Extensive experiments demonstrate that our method simultaneously generates corresponding videos and motions while performing accurate motion capture. This work taps into the capacity of LLMs to fuse diverse data types, paving the way for integrating human-centric information into existing models and potentially enabling multimodal, controllable joint modeling of humans, objects, and scenes. ---