8771 papers
This paper introduces MSVBench, the first comprehensive benchmark designed to evaluate multi-shot video generation through hierarchical scripts and a hybrid LMM-expert framework, revealing that current models lack true world modeling capabilities while achieving near-perfect alignment with human judgments.
This paper introduces Venus, a two-stage framework built upon the new AesGuide dataset, which empowers multimodal large language models to provide actionable aesthetic guidance and achieve state-of-the-art performance in aesthetic cropping by addressing the gap between ordinary users and professional photographers.
The paper proposes MINT, a novel three-stage framework that transfers biomarker knowledge from structural MRI to speech analysis during training, enabling biologically grounded, non-invasive early Alzheimer's screening at the population scale without requiring neuroimaging at inference.
The paper proposes MIGM-Shortcut, a lightweight method that learns latent controlled dynamics by regressing feature evolution velocities using both previous features and sampled tokens, achieving over 4x acceleration in masked image generation while maintaining high quality.
This paper proposes an ordinal latent diffusion model that leverages the continuous, ordered nature of diabetic retinopathy severity to generate more realistic and clinically consistent color fundus images compared to standard categorical conditional diffusion models.
This paper introduces DeBiasLens, an interpretable, model-agnostic framework that utilizes sparse autoencoders to identify and selectively deactivate social attribute neurons within Vision-Language models, thereby mitigating social biases without compromising semantic knowledge.
The paper proposes SR3R, a feed-forward framework that reformulates 3D super-resolution as a direct mapping from sparse low-resolution views to high-resolution 3D Gaussian Splatting representations, enabling robust zero-shot generalization and superior reconstruction fidelity by autonomously learning 3D-specific high-frequency details from large-scale multi-scene data.
This paper proposes SteerVAD, a novel tuning-free framework that enhances video anomaly detection in frozen multi-modal LLMs by identifying latent anomaly experts and employing a hierarchical meta-controller to dynamically steer and rectify their internal representations, thereby achieving state-of-the-art performance with minimal training data.
GuardAlign is a training-free defense framework for multimodal large language models that combines optimal transport-based safety detection and cross-modal attentive calibration to significantly reduce unsafe response rates while preserving model utility.
This paper proposes Adaptive Visual Reinforcement (AIR), a training-free framework that mitigates hallucinations in Multimodal Large Language Models by condensing visual tokens and selectively reinforcing the most consistent image patches to enhance reliance on salient visual evidence.
This paper presents a unified framework for high-fidelity 3D garment reconstruction from monocular images and videos by combining Implicit Sewing Patterns with a generative diffusion model in UV space to learn expressive shape priors and enforce spatio-temporal consistency, enabling accurate recovery of both tight- and loose-fitting clothing with fine geometric details.
This paper proposes Quant Experts (QE), a token-aware adaptive error reconstruction method using a mixture of shared and routed low-rank adapters to dynamically compensate for quantization errors in Large Vision-Language Models, thereby achieving near full-precision performance across various scales without retraining.
This paper introduces a neurosymbolic verification framework that uses SMT solvers and clinical knowledge bases to formally audit and guarantee the logical consistency of vision-language model-generated radiology reports, effectively identifying and eliminating hallucinations and deductive failures that traditional metrics miss.
AgenticOCR introduces a dynamic, query-driven parsing paradigm that autonomously identifies and extracts only relevant document regions to overcome the context overload and hallucination risks of traditional page-level chunking in multimodal Retrieval-Augmented Generation systems.
This paper proposes an efficient 3D Gaussian Splatting framework that combines an adaptive, reconstruction-aware pruning strategy with novel Difference-of-Gaussians primitives to achieve up to 90% model compression while maintaining or enhancing rendering quality.
This paper introduces TASOT, an unsupervised multimodal optimal transport framework that leverages visual and text-based cues to achieve state-of-the-art surgical phase and step segmentation without relying on costly large-scale pre-training.
HumanOrbit is a video diffusion-based method that generates consistent 360° orbit videos from a single human image, enabling the reconstruction of high-fidelity, geometrically complete 3D textured meshes with superior identity preservation compared to existing multi-view synthesis approaches.
The paper proposes RAViT, a resolution-adaptive vision transformer framework that utilizes a multi-branch architecture with an early exit mechanism to achieve accuracy comparable to standard Vision Transformers while significantly reducing computational costs by processing images at varying resolutions.
This paper introduces a novel superpixel hierarchy for high-dimensional images that integrates spatial layout with attribute manifold information, enabling consistent and effective exploration of both image and attribute spaces compared to traditional methods that ignore spatial coherence.
The paper proposes the Omnivorous Vision Encoder, a framework that aligns feature representations across diverse modalities (such as RGB and depth) while distilling knowledge from a frozen DINOv2 teacher, thereby enabling robust, modality-agnostic scene understanding.