Ambiguous 3D medical image segmentation often involves boundaries where different expert delineations are non-identical yet clinically plausible. Modeling such inter-observer variability requires a careful balance between diversity and anatomical fidelity: deterministic models preserve coherent volumetric structures but collapse expert disagreement into a single mask, while stochastic generative models can produce diverse samples but may introduce disconnected components or slice-to-slice inconsistency when generating full 3D masks from unstructured noise. We propose Volumetric Directional Diffusion (VDD), a prior-anchored diffusion framework that shifts stochastic generation from full-mask synthesis to residual boundary exploration. VDD uses a coarse consensus prediction as an anatomical anchor and learns a directional diffusion process to generate plausible boundary variations around ambiguous regions while preserving stable volumetric topology. Experiments on three multi-rater datasets, including LIDC-IDRI, KiTS21, and ISBI 2015, show that VDD improves uncertainty distribution alignment while maintaining competitive segmentation accuracy and 3D structural consistency. These results suggest that prior-anchored residual diffusion can model clinically plausible expert disagreement without sacrificing anatomical fidelity.

Diffusion models have emerged as the backbone of modern generative AI, powering advances in vision, language, audio and other modalities. Despite their success, they suffer from hallucinations, implausible samples that lie outside the support of true data distribution, which degrade reliability and trust. In this work, we first empirically confirm previously proposed hypothesis that score smoothness causes hallucinations in Image Generation diffusion models and provide a density-based perspective. We further formalize this notion by linking the hallucinations probability mass to lipschitz constant of the learned score function. Motivated by this, we introduce a Variance-Guided Score Modulation (VSM) strategy that controls the score Jacobian, in turn reducing score smoothness and better approximating the ground truth score that decreases hallucinations. Empirical results on synthetic and real-world datasets demonstrate that our approach reduces hallucinations (up to 25

Textured high-fidelity 3D models are crucial for games, AR/VR, and film, but human-aligned evaluation methods still fall behind despite recent advances in 3D reconstruction and generation. Existing metrics, such as Chamfer Distance, often fail to align with how humans evaluate the fidelity of 3D shapes. Recent learning-based metrics attempt to improve this by relying on rendered images and 2D image quality metrics. However, these approaches face limitations due to incomplete structural coverage and sensitivity to viewpoint choices. Moreover, most methods are trained on synthetic distortions, which differ significantly from real-world distortions, resulting in a domain gap. To address these challenges, we propose a new fidelity evaluation method that is based directly on 3D meshes with texture, without relying on rendering. Our method, named Textured Geometry Evaluation TGE, jointly uses the geometry and color information to calculate the fidelity of the input textured mesh with comparison to a reference colored shape. To train and evaluate our metric, we design a human-annotated dataset with real-world distortions. Experiments show that TGE outperforms rendering-based and geometry-only methods on real-world distortion dataset.

Grounded multi-video question answering over real-world news events requires systems to surface query-relevant evidence across heterogeneous video archives while attributing every claim to its supporting source. We introduce CRAFT (Critic-Refined Adaptive Key-Frame Targeting), a query-conditioned pipeline that combines dynamic keyframe selection, per-video ASR with multilingual fallback, and a hybrid critic loop to iteratively verify and repair claims before consolidation. The pipeline integrates UNLI temporal entailment, DeBERTa-v3 cross-claim screening, and a Llama-3.2-3B adjudicator, with a final citation-merging stage that emits each fact once with all supporting source identifiers. On MAGMaR 2026, CRAFT achieves the best overall average (0.739), reference recall (0.810), and citation F1 (0.635). We further evaluate on a MAGMaR-style conversion of WikiVideo with 52 non-overlapping event queries, where CRAFT also performs strongly (0.823 Avg), showing that its claim-centric evidence aggregation generalizes beyond MAGMaR. Ablations show that atomic claims, ASR, and the critic loop drive the main gains over the vanilla query-conditioned baseline. Code and implementation details are publicly available at https://github.com/bhosalems/CRAFT.

Multi-video event understanding demands models that can locate and attribute query-relevant evidence scattered across long, heterogeneous video corpora. Existing large vision-language models (LVLMs) often underperform in this regime because they quickly exhaust their context budget and struggle to precisely localize evidentially important segments, frequently missing dense informational cues such as broadcast graphics, subtitles, and scoreboards. We introduce TRACE, an evidence grounding-guided framework that follows a ground-before-reasoning strategy for multi-video event reasoning. Our approach first builds a structured, text-searchable timeline for each video using OCR and object detection. A text-only LLM then conducts query-aware evidence localization, selecting relevant moments prior to any downstream visual reasoning. The retrieved frames and their grounding summaries are subsequently used to steer LVLM-based claim generation and cross-video citation consolidation. Experiments on MAGMaR 2026 and WikiVideo demonstrate that structured grounding markedly boosts factual completeness and attribution fidelity. On the MAGMaR validation split, TRACE raises macro-average MiRAGE F1 from 0.705 to 0.811 compared to an unguided Qwen3-VL-30B baseline, with especially strong improvements in citation recall from 0.440 to 0.628. The method also attains state-of-the-art results on the official MAGMaR 2026 leaderboard.

Interpreting chest X-rays is inherently challenging due to the overlap between anatomical structures and the subtle presentation of many clinically significant pathologies, making accurate diagnosis time-consuming even for experienced radiologists. Recent radiology-focused foundation models, such as LLaVA-Rad and Maira-2, have positioned multi-modal large language models (MLLMs) at the forefront of automated radiology report generation (RRG). However, despite these advances, current foundation models generate reports in a single forward pass. This decoding strategy diminishes attention to visual tokens and increases reliance on language priors as generation proceeds, which in turn introduces spurious pathology co-occurrences in the generated reports. To mitigate these limitations, we propose Category-Wise Contrastive Decoding (CWCD), a novel and modular framework designed to enhance structured radiology report generation (SRRG). Our approach introduces category-specific parameterization and generates category-wise reports by contrasting normal X-rays with masked X-rays using category-specific visual prompts. Experimental results demonstrate that CWCD consistently outperforms baseline methods across both clinical efficacy and natural language generation metrics. An ablation study further elucidates the contribution of each architectural component to overall performance.

Referring Image Segmentation (RIS) aims to segment image regions specified by natural language, enabling fine-grained and controllable visual understanding. Extending RIS to endoscopic imagery, however, presents unique challenges, including scarce high-quality annotations and complex, domain-specific image-text relationships. Although recent vision-language models demonstrate strong cross-domain alignment, they often fail to capture fine-grained textual cues in endoscopic settings, resulting in suboptimal performance and limited generalization. To address these challenges, we introduce ReferEndoscopy, a large-scale benchmark for RIS in the endoscopy field. Building on this dataset, we propose the Attribute Retrieval-based Endoscopic-RIS (AR-ERIS) framework for open-vocabulary endoscopic compositional referring segmentation. AR-ERIS leverages attribute retrieval for open-vocabulary endoscopic compositional referring segmentation and is pretrained on the curated ReferEndoscopy dataset, achieving state-of-the-art performance with strong generalization across both simulated and real-world endoscopic data. The dataset and code will be publicly released upon completion of the review process.

Deep networks often rely on architectural heuristics to shape representation evolution, limiting their ability to model data governed by intrinsic dynamics. We present the Circuit-inspired High-Order Neural Network (CHONN), a modular framework that treats representation evolution as a latent potential process and increases its effective order through Kirchhoff-inspired cascade composition. A single Kirchhoff Neural Cell implements a stable first-order update, while serially composed cells form higher-order dynamical operators within one block. This construction is interpretable, numerically stable and compatible with common neural backbones. Theoretical analysis shows that cascaded cells induce end-to-end high-order operators, and controlled experiments demonstrate that intra-block high-order construction differs from generic depth stacking, especially on derivative-sensitive measures. Across steady-state operator learning, long-horizon physical forecasting and ImageNet-1K recognition, CHONN improves structural fidelity, rollout stability and visual representation learning. These results identify high-order circuit composition as a general principle for neural dynamics modeling.

This survey presents a comprehensive overview of recent advances in Handwritten Text Generation (HTG), focusing on prominent model architectures such as Generative Adversarial Networks (GANs), Transformer-based models, and Diffusion Models. We analyze their respective strengths and limitations in terms of training stability, style transfer, and output fidelity. In addition, we discuss a range of modeling strategies — from sequence modeling to style-content disentanglement. We also expose persistent challenges related to model training, evaluation metrics, and dataset scarcity. By critically discussing these methodologies, we aim to provide information on state-of-the-art practices and emerging trends, offering a resource for researchers and practitioners seeking to advance the field of handwriting generation.

In computer vision, the challenging task of Visual Question Answering (VQA) requires multi-modal reasoning to understand both visual content and linguistic context. The current state-of-the-art methods for document understanding either analyze entire table images or use patch-based retrieval. In this paper, we propose MT-VQA, an efficient VQA method for tabular data that is both globally and locally cell-content-aware. We first extract a segmentation map from each table using a structure detector, storing headers and row labels as JSON, used to build a relational graph, then classifying whether a question needs one or more tables, to perform keyword matching or multi-hop graph retrieval. A region selector selects regions relevant to the query, followed by context assembler for building the final prompt. We evaluated MT-VQA zero-shot on ComTQA and TableVQA-Bench, achieving the SOTA accuracy of 54\% on ComTQA and average accuracy of 66.3\% on TableVQA-Bench. Based on our findings, the combination of structural map routing with graph-based cross-table reasoning is a promising approach for Visual Table Question Answering.

While powerful in image-conditioned generation, multimodal large language models (MLLMs) can display uneven performance across demographic groups, highlighting fairness risks. In safety-critical clinical settings, such disparities risk producing unequal diagnostic narratives and eroding trust in AI-assisted decision-making. While fairness has been studied extensively in vision-only and language-only models, its impact on MLLMs remains largely underexplored. To address these biases, we introduce FairLLaVA, a parameter-efficient fine-tuning method that mitigates group disparities in visual instruction tuning without compromising overall performance. By minimizing the mutual information between target attributes, FairLLaVA regularizes the model’s representations to be demographic-invariant. The method can be incorporated as a lightweight plug-in, maintaining efficiency with low-rank adapter fine-tuning, and provides an architecture-agnostic approach to fair visual instruction following. Extensive experiments on large-scale chest radiology report–generation task shows FairLLaVA consistently reduces inter-group gaps while improving equity-scaled clinical performance and natural language generation metrics. Code and models will be open-sourced.

Recent advancements in text-to-image (T2I) generation have revolutionized image synthesis, but conventional text-image paired training poses challenges when confronted with limited dataset size and narrow descriptive breadth. In particular, limited descriptive breadth can significantly impair a model’s ability to generate unmentioned image features. This issue is also evident in text-to-face image generation, where a method will underperform when rendering an image based on the text description of a specific facial feature not present in the training dataset. Language-free training emerges as a promising solution to this problem, leveraging latent spaces like CLIP to facilitate generalization from image embeddings to text embeddings. However, the modality gap remains a hurdle for language-free trained models. To address this, we propose a Gaussian perturbation-based technique that enhances perturbation coverage and robustness across varying modality gap sizes without the need to train a prior model. Our method achieves new state-of-the-art results on MM-CelebA-HQ in the language-free setting, presenting a novel solution to challenges in text-to-face image generation on limited datasets.

The personalization model has gained significant attention in image generation yet remains underexplored for large vision-language models (LVLMs). Beyond generic ones, with personalization, LVLMs handle interactive dialogues using referential concepts (e.g., “Mike and Susan are talking.”) instead of the generic form (e.g., “a boy and a girl are talking.”), making the conversation more customizable and referentially friendly. In addition, PLVM is equipped to continuously add new concepts during a dialogue without incurring additional costs, which significantly enhances the practicality. PLVM proposes Aligner, a pre-trained visual encoder to align referential concepts with the queried images. During the dialogues, it extracts features of reference images with these corresponding concepts and recognizes them in the queried image, enabling personalization. We note that the computational cost and parameter count of the Aligner are negligible within the entire framework. With comprehensive qualitative and quantitative analyses, we reveal the effectiveness and superiority of PLVM.

Feature matching methods for unsupervised anomaly detection have demonstrated impressive performance. Existing methods primarily rely on self-supervised training and handcrafted matching schemes for task adaptation. However, they can only achieve an inferior feature representation for anomaly detection because the feature extraction and matching modules are separately trained. To address these issues, we propose a Differentiable Feature Matching (DFM) framework for joint optimization of the feature extractor and the matching head. DFM transforms nearest-neighbor matching into a pooling-based module and embeds it within a Feature Matching Network (FMN). This design enables end-to-end feature extraction and feature matching module training, thus providing better feature representation for anomaly detection tasks. DFM is generic and can be incorporated into existing feature-matching methods. We implement DFM with various backbones and conduct extensive experiments across various tasks and datasets, demonstrating its effectiveness. Notably, we achieve state-of-the-art results in the continual anomaly detection task with instance-AUROC improvement of up to 3.9% and pixel-AP improvement of up to 5.5%.

Quantifying the causal influence of input features within neural networks has become a topic of increasing interest. Existing approaches typically assess direct, indirect, and total causal effects. This work treats NNs as structural causal models (SCMs) and extends our focus to include intrinsic causal contributions (ICC). We propose an identifiable generative post-hoc framework for quantifying ICC. We also draw a relationship between ICC and Sobol' indices. Our experiments on synthetic and real-world datasets demonstrate that ICC generates more intuitive and reliable explanations compared to existing global explanation techniques.

Diffusion-based generative models have shown promise in synthesizing histopathology images to address data scarcity caused by privacy constraints. Diagnostic text reports provide high-level semantic descriptions, and masks offer fine-grained spatial structures essential for representing distinct morphological regions. However, public datasets lack paired text and mask data for the same histopathological images, limiting their joint use in image generation. This constraint restricts the ability to fully exploit the benefits of combining both modalities for enhanced control over semantics and spatial details. To overcome this, we propose PathDiff, a diffusion framework that effectively learns from unpaired mask-text data by integrating both modalities into a unified conditioning space. PathDiff allows precise control over structural and contextual features, generating high-quality, semantically accurate images. PathDiff also improves image fidelity, text-image alignment, and faithfulness, enhancing data augmentation for downstream tasks like nuclei segmentation and classification. Extensive experiments demonstrate its superiority over existing methods.

Existing outfit recommendation frameworks focus on outfit compatibility prediction and complementary item retrieval. We present a text-driven outfit generation framework, Text2Outfit, which generates outfits controlled by text prompts. Our framework supports two forms of outfit recommendation: 1) Text-to-outfit generation, where the prompt includes the specification for each outfit item (e.g., product features), and the model retrieves items that match the prompt and are stylistically compatible. 2) Seed-to-outfit generation, where the prompt includes the specification for a seed item, and the model both predicts which product types the outfit should include (referred to as composition generation) and retrieves the remaining items to build an outfit. We develop a large language model (LLM) framework that learns the cross-modal mapping between text and image set, and predicts a set of embeddings and compositions to retrieve outfit items. We devise an attention masking mechanism in LLM to handle the alignment between text descriptions and image tokens from different categories. We conduct experiments on the Polyvore dataset and evaluate the quality of the generated outfits from two perspectives: 1) feature matching for outfit items, and 2) outfit visual compatibility. The results demonstrate that our approach significantly outperforms the baseline methods in text to outfit generation.

Recent advances in diffusion models have revolutionized text-guided image editing, yet existing editing methods face critical challenges in hyperparameter identification. To get the reasonable editing performance, these methods often require the user to brute-force tune multiple interdependent hyperparameters, such as inversion timesteps and attention modification, \textit{etc.} This process incurs high computational costs due to the huge hyperparameter search space. We consider searching optimal editing's hyperparameters as a sequential decision-making task within the diffusion denoising process. Specifically, we propose a reinforcement learning framework, which establishes a Markov Decision Process that dynamically adjusts hyperparameters across denoising steps, integrating editing objectives into a reward function. The method achieves time efficiency through proximal policy optimization while maintaining optimal hyperparameter configurations. Experiments demonstrate significant reduction in search time and computational overhead compared to existing brute-force approaches, advancing the practical deployment of a diffusion-based image editing framework in the real world.

Human motion prediction (HMP) forecasts future human motion (pose sequences) based on previous motion data. While existing methods excel by learning motion dynamics from adjacent 3D skeleton poses, they face a significant challenge: the reliance on large-scale 3D pose annotations, which are costly to produce and often fail to capture the full diversity of actions and scenarios. This issue is particularly pronounced for underrepresented groups, such as the elderly, where annotated 3D data is even scarcer. To address this challenge, we propose leveraging more readily available 2D annotations to complement the limited 3D data for HMP. In this work, we introduce UniMotion, a unified system for HMP capable of predicting both 2D and 3D future human pose sequences from either 2D and/or 3D previous pose sequences. The main advantage of UniMotion with respect to previous HMP systems is that it requires much less 3D training data, obtaining remarkable accuracies even when trained with just a small portion of 3D data. To train UniMotion with unpaired 2D and 3D pose sequences, we introduce a novel sequential bidirectional knowledge distillation module (SeqBi), which enables mutual learning between the 2D and 3D encoders. To tackle the data imbalance challenge, we increase the diversity of the underrepresented 3D data by adding a small perturbation to the joint angles at the sequence level (RegPer). Extensive experiments on public datasets, including general adult datasets (H3.6M, 3DPW) and an elderly-specific dataset (TST), demonstrate that UniMotion achieves results comparable to or better than state-of-the-art methods while requiring only one-third of the 3D training data.

Enhancing the network architecture of the YOLO framework has been crucial for a long time. Still, it has focused on CNN-based improvements despite the proven superiority of attention mechanisms in modeling capabilities. This is because attention-based models cannot match the speed of CNN-based models. This paper proposes an attention-centric YOLO framework, namely YOLOv12, that matches the speed of previous CNN-based ones while harnessing the performance benefits of attention mechanisms. YOLOv12 surpasses popular real-time object detectors in accuracy with competitive speed. For example, YOLOv12-N achieves 40.5% mAP with an inference latency of 1.62 ms on a T4 GPU, outperforming advanced YOLOv10-N / YOLO11-N by 2.0%/1.1% mAP with a comparable speed. This advantage extends to other model scales. YOLOv12 also surpasses end-to-end real-time detectors that improve DETR, such as RT-DETRv2 / RT-DETRv3: YOLOv12-X beats RT-DETRv2-R101 / RT-DETRv3-R101 while running faster with fewer computations and parameters. See more comparisons in Figure 1. Source code is available at https://github.com/sunsmarterjie/yolov12.