5504 papers
This paper proposes a memory-efficient unsupervised anomaly detection framework that leverages a 2D autoregressive CNN to explicitly model spatial dependencies in DINOv3 patch embeddings, achieving competitive performance on medical imaging benchmarks while significantly reducing inference time and memory overhead compared to existing prototype-based methods.
The Dresden Dataset (D4D) is a comprehensive benchmark comprising over 300,000 frames and 369 point clouds from porcine cadaver surgeries, providing paired endoscopic video and high-quality structured-light geometry to enable quantitative evaluation of non-rigid 4D reconstruction, SLAM, and depth estimation methods in realistic abdominal surgical scenes.
VIRGi is a novel method for rapidly and photorealistically recoloring 3D Gaussian Splatting scenes by decomposing color into diffuse and view-dependent components, enabling real-time, single-image editing that preserves specular highlights and outperforms NeRF-based approaches.
The paper proposes the Ultra Resolution Geometry Transformer (URGT), a unified multi-patch framework that leverages cross-patch attention and a GridMix sampling strategy to achieve state-of-the-art, high-resolution monocular depth and normal estimation with enhanced global consistency and generalization.
The paper introduces BRIGHT, a collaborative generalist-specialist foundation model trained on a massive breast-specific dataset that outperforms existing generalist models across 24 diverse clinical tasks, establishing a new state-of-the-art for breast pathology and validating a scalable paradigm for organ-specific AI development.
This paper introduces EduAIGV-1k, the first benchmark dataset for evaluating AI-generated educational videos on math concepts, and proposes EduVQA, a novel assessment framework featuring a Structured 2D Mixture-of-Experts module that leverages fine-grained perceptual and prompt-alignment annotations to outperform existing video quality baselines.
The paper introduces TikZilla, a family of small open-source models trained on the high-quality DaTikZ-V4 dataset using a two-stage pipeline of supervised fine-tuning and reinforcement learning with image-based rewards, which significantly outperforms prior approaches and matches GPT-5 in generating accurate TikZ figures from text descriptions.
This paper introduces TinyIceNet, a compact, low-power semantic segmentation network co-designed for FPGA deployment that enables near-real-time, energy-efficient sea ice mapping directly on-board satellites using Sentinel-1 SAR imagery.
The paper proposes AWDiff, a novel diffusion-based framework that integrates the a trous wavelet transform and BioMedCLIP semantic conditioning to generate high-fidelity lung ultrasound images while preserving critical fine-scale diagnostic structures often lost in existing methods.
This paper proposes RL3DEdit, a reinforcement learning framework that leverages 3D foundation model priors from VGGT to generate novel rewards for optimizing 2D diffusion-based editing, thereby achieving stable multi-view consistency in 3D scene editing without requiring paired training data.
Kling-MotionControl is a unified DiT-based framework that achieves robust, high-fidelity holistic character animation by orchestrating specialized motion representations for body, face, and hands while ensuring cross-identity generalization, faithful appearance preservation, and over 10x faster inference through advanced distillation.
To address the trade-off between safety and image quality in Text-to-Image models, the authors propose Conditioned Activation Transport (CAT), a framework that utilizes a contrastive dataset and geometry-based nonlinear transport maps to steer unsafe activations without degrading benign generations.
This paper proposes MoD-DPO, a Modality-Decoupled Direct Preference Optimization framework that mitigates cross-modal hallucinations in omni-modal LLMs by enforcing modality-specific invariance and sensitivity through regularization and language-prior debiasing, thereby significantly improving perception accuracy and hallucination resistance.
This paper introduces CoWVLA, a novel Vision-Language-Action framework that unifies world-model temporal reasoning with disentangled latent motion representations to enable efficient and accurate visuomotor learning by predicting continuous latent motion chains and aligning them with discrete action sequences.
This paper introduces ACE-Brain-0, a generalist foundation brain that leverages spatial intelligence as a universal scaffold and employs a Scaffold-Specialize-Reconcile (SSR) paradigm to unify diverse embodied tasks like autonomous driving and robotics within a single multimodal large language model, achieving state-of-the-art performance across 24 benchmarks.
COP-GEN is a multimodal latent diffusion transformer designed for Earth observation that addresses the inherent non-injectivity of cross-sensor relationships by modeling conditional distributions to generate diverse, physically consistent, and uncertainty-aware realizations across optical, radar, and elevation modalities without task-specific retraining.
The paper introduces UniG2U-Bench, a comprehensive benchmark evaluating generation-to-understanding across 30 subtasks, revealing that while unified multimodal models generally underperform base vision-language models, they offer specific advantages in spatial intelligence and multi-round reasoning, thereby highlighting the need for improved training data and paradigms.
DuoMo is a state-of-the-art generative method that reconstructs globally consistent, world-space human motion from unconstrained and noisy videos by employing a dual diffusion framework that first estimates camera-space motion and then refines it into a coherent world-space trajectory without relying on parametric models.
LoGeR introduces a novel architecture featuring a hybrid memory module that combines parametric Test-Time Training and non-parametric Sliding Window Attention to enable dense 3D geometric reconstruction of extremely long video sequences with global consistency, significantly outperforming existing feedforward methods without requiring post-optimization.
This paper presents controlled from-scratch pretraining experiments using the Transfusion framework to demonstrate that a Mixture-of-Experts architecture effectively harmonizes the data-scaling asymmetry between vision and language, enabling unified multimodal models with superior world modeling and complementary capabilities.