4014 papers
The paper proposes AR2-4FV, a novel framework for long-term language-guided referring in fixed-view videos that leverages a static background-derived Anchor Bank and a ReID-Gating mechanism to maintain identity continuity and accelerate re-capture during occlusions or scene exits, significantly outperforming existing baselines in re-capture rate and latency.
The paper proposes DECADE, an unsupervised diffusion model that achieves temporally consistent denoising of Rb-82 dynamic cardiac PET images without paired training data, effectively reducing noise while preserving quantitative accuracy for myocardial blood flow and flow reserve metrics.
This paper introduces MedQ-Deg, a comprehensive benchmark featuring 24,894 expert-calibrated question-answer pairs across 18 degradation types and 7 imaging modalities, which reveals that mainstream medical multimodal large language models suffer systematic performance drops and exhibit the "AI Dunning-Kruger Effect" of overconfidence under image quality degradations.
The paper introduces Apply2Isar, a tool that automatically converts procedural apply-style proofs in Isabelle/HOL into readable and robust declarative Isar proofs, demonstrating its effectiveness through evaluation on a large benchmark set from the Isabelle Archive of Formal Proofs.
This paper proposes the Geometric Knowledge-Guided Federated Dual Knowledge Distillation (GK-FedDKD) framework to address data heterogeneity in remote sensing satellite imagery by leveraging aggregated geometric knowledge from local covariance matrices and a dual distillation process to significantly outperform state-of-the-art methods.
This paper proposes a stability-aligned residual control architecture that enables robotic systems to rapidly recover from mid-episode dynamics shifts by keeping the nominal policy frozen while using a bounded, gated additive residual channel to adaptively compensate for unobserved disturbances, achieving up to an 87% reduction in recovery time across various robotic platforms.
This paper proposes a novel style transfer method that represents images in the brushstroke domain rather than the traditional pixel domain, offering a more natural and visually superior approach to mimicking artistic styles.
The paper introduces OrdinalBench, a comprehensive benchmark dataset and evaluation framework designed to diagnose and expose the significant generalization limitations of Vision-Language Models in understanding ordinal numbers and performing sequential reasoning tasks involving large indices and complex paths.
The paper proposes Structured Gaussian Image (SGI), a framework that represents high-resolution images using multi-scale, seed-based structured 2D Gaussians generated by lightweight MLPs, achieving significant compression and faster convergence compared to existing unstructured 2D Gaussian methods while maintaining high image fidelity.
This paper introduces 4DRC-OCC, the first framework to fuse 4D radar and camera data for robust 3D semantic occupancy prediction, leveraging their complementary strengths to overcome adverse weather and lighting challenges while utilizing a newly created automatically labeled dataset to reduce annotation costs.
This paper presents a multi-legged robot equipped with biomimetic, gradient-compliant tactile antennae that enable robust navigation and recovery in confined, obstacle-rich environments by mapping antenna deformation to collision states for real-time steering without relying on global environmental information or vision.
This paper introduces Inverse Resistive Force Theory (I-RFT), a physics-informed machine learning framework that enables robots to accurately estimate granular terrain properties from proprioceptive contact forces under arbitrary gait trajectories, thereby facilitating data-efficient environmental characterization and adaptive locomotion strategies.
This paper introduces MWM, a mobile world model that enhances action-conditioned rollout consistency and inference efficiency for image-goal navigation through a novel two-stage training framework featuring Action-Conditioned Consistency post-training and Inference-Consistent State Distillation.
The paper introduces STEP, a preference-conditioned reinforcement learning framework that optimizes robotic 3D bin packing by explicitly balancing spatial efficiency against operational time, achieving a 44% reduction in execution time without compromising packing density.
This paper investigates the non-injectivity of the verbose persistent homology transform (VPHT) for graphs with collinear vertices, identifying necessary and sufficient conditions for their non-reconstructibility to advance a complete classification of such cases.
This paper proposes a temperature-aware, distributed optimization approach using the alternating direction method of multipliers to co-optimize energy, carbon, water, and latency costs for LLM inference across geo-distributed edge data centers in Australia, leveraging ambient temperature diversity to enhance sustainability and efficiency.
This paper examines the governance of AI-generated content across 40 social media platforms, finding that while most focus on moderation and disclosure, few address ownership and monetization, prompting a call for more comprehensive policies and user education.
HybridStitch is a novel Text-to-Image generation paradigm that accelerates diffusion models by treating generation as an editing process, dynamically splitting the image into easy and complex regions to leverage a small model for coarse sketching and a large model for targeted refinement, thereby achieving a 1.83× speedup on Stable Diffusion 3.
This study demonstrates that low-cost, animal-triggered camera traps combined with foundation vision models can effectively monitor fine-grained plant phenology and flora-faunal interactions in a Hawaiian cloud forest, revealing ecological trends that traditional sampling methods often miss.
This paper presents a curved monopole antenna design optimized for HF skywave radar that, through parametric analysis of curvature and straight-section length, achieves significantly enhanced gain and bandwidth compared to conventional monopoles, and demonstrates further performance improvements when scaled into a 12-element linear array.