4144 papers
GEM-TFL is a novel framework that bridges the performance gap between weakly and fully supervised temporal forgery localization by employing EM-guided label decomposition, graph-based proposal refinement, and training-free temporal consistency to overcome the limitations of binary video-level supervision.
Diff-ES is a novel framework that leverages evolutionary search to automatically optimize stage-wise sparsity schedules for diffusion models, enabling memory-efficient structural pruning that achieves significant speedups with minimal quality degradation.
This paper introduces BLINK, a trajectory-based recurrent state-space model that learns latent interaction dynamics from time-lapse microscopy to accurately predict and forecast NK cell cytotoxic outcomes while providing an interpretable representation of cellular behavioral modes.
UniPAR is a unified Transformer-based framework that overcomes the limitations of existing "one-model-per-dataset" approaches by integrating a phased fusion encoder and dynamic scheduling to simultaneously process heterogeneous modalities (RGB, video, and event streams) across diverse datasets, achieving state-of-the-art performance and enhanced cross-domain robustness.
The paper proposes SRasP, a novel method for Cross-Domain Few-Shot Learning that mitigates domain shift and improves generalization by using global semantic guidance to reorient and aggregate style gradients, thereby stabilizing training and encouraging convergence to flatter, more transferable solutions.
This paper proposes a complexity-aware adaptive inference framework for Vision-Language-Action models that dynamically routes execution to "Act," "Think," or "Abstain" based on task complexity, leveraging a vision-only detector to optimize resource allocation and prevent failures while achieving high accuracy with minimal training data.
This paper proposes SSR-GS, a novel framework that enhances glossy surface reconstruction in 3D Gaussian Splatting by modeling direct and indirect specular reflections through a prefiltered Mip-Cubemap and an IndiASG module, while employing Visual Geometry Priors to mitigate the impact of reflections on photometric loss.
This study demonstrates that simple lung cropping preprocessing effectively mitigates racial shortcut learning in chest X-ray diagnosis models by suppressing spurious racial cues while preserving diagnostic accuracy, thereby avoiding the typical fairness-accuracy trade-off.
This paper proposes a method for generic camera calibration using blurry images by simultaneously estimating feature locations and spatially varying point spread functions through geometric constraints and a local parametric illumination model, thereby overcoming the motion blur issues inherent in capturing numerous calibration images.
The paper proposes Mario, a unified framework that enhances large language model-based reasoning on multimodal graphs by employing a graph-conditioned vision-language model for cross-modal feature refinement and a modality-adaptive instruction tuning mechanism to dynamically select optimal modality configurations, thereby outperforming existing state-of-the-art methods in node classification and link prediction tasks.
Logi-PAR is a novel framework that enhances patient activity recognition in clinical settings by integrating contextual fact fusion and learnable differentiable logic rules to provide auditable explanations and counterfactual interventions, achieving state-of-the-art performance on benchmarks like VAST and OmniFall.
The paper proposes the Semantic Class Distribution Learning (SCDL) framework, a plug-and-play module that mitigates supervision and representation biases in semi-supervised medical image segmentation by learning structured class-conditional feature distributions, thereby achieving state-of-the-art performance with significant improvements on minority classes.
The paper introduces SPyCer, a semi-supervised physics-guided deep learning framework that leverages satellite imagery and physical constraints derived from surface energy balance and advection-diffusion-reaction equations to generate accurate, spatially continuous estimates of near-surface air temperature, outperforming existing methods in both accuracy and physical consistency.
This paper presents a digital twin-driven dual-arm robotic system that integrates RGBD sensing, tactile feedback, and state-of-the-art Visual Language Models to achieve robust, real-time textile classification and foreign object detection for automated sustainable recycling.
The paper introduces ICHOR, a self-supervised masked autoencoder framework pretrained on a large, multi-site dataset of 11,405 ASL CBF maps, which outperforms existing methods in learning robust, transferable representations for downstream diagnostic and quality prediction tasks.
CATNet is a collaborative perception framework that addresses real-world challenges of temporal latency and multi-source noise through its novel Spatio-Temporal Recurrent Synchronization, Dual-Branch Wavelet Enhanced Denoiser, and Adaptive Feature Selector, demonstrating superior robustness and adaptability in complex traffic conditions.
The paper proposes Wiki-R1, a curriculum reinforcement learning framework that employs controllable data generation and a strategic sampling method to systematically bridge the distributional gap between pretrained multimodal models and knowledge-based VQA tasks, achieving state-of-the-art performance on Encyclopedic VQA and InfoSeek benchmarks.
This paper demonstrates that distribution shift is the primary cause of performance degradation in deeper layers of Vision Transformers and reveals that optimal out-of-distribution probing requires selecting between feedforward network activations and normalized self-attention outputs depending on the severity of the shift.
The paper introduces WebChain, a large-scale, human-annotated dataset of real-world web interaction traces featuring multi-modal alignment, which enables a novel dual mid-training approach that achieves state-of-the-art performance in web agent planning and spatial grounding.
Fusion4CA is a plug-and-play framework that enhances 3D object detection by fully exploiting RGB information through a contrastive alignment module, a camera auxiliary branch, and cognitive adapters, achieving significant performance gains with minimal parameter overhead and superior training efficiency.