5561 papers
The paper introduces SSR, a 7B-parameter framework that achieves state-of-the-art spatial intelligence by integrating 2D and 3D representations through lightweight alignment and a novel scene graph generation pipeline, enabling precise geometric reasoning without costly large-scale pre-training.
To overcome the scarcity of 3D-text data and the resulting loss of geometric information in existing 3D Vision-Language Models, PointAlign introduces a lightweight feature-level alignment regularization that explicitly supervises intermediate point cloud tokens to preserve fine-grained 3D geometric-semantic details, significantly improving performance on classification and captioning tasks.
This paper presents DiffTrans, a differentiable rendering framework that utilizes FlexiCubes for initial geometry and a recursive CUDA-based ray tracer to jointly optimize geometry, refractive index, and absorption, enabling high-quality reconstruction of transparent objects with diverse topologies and complex textures in intricate scenes.
The paper proposes Query-Conditioned Gaussian Splatting (QCGS), a novel framework that fuses sparse weather station data with satellite imagery to efficiently generate high-resolution precipitation fields by selectively rendering only rainfall regions, achieving over 50% improvement in RMSE compared to conventional products.
This paper introduces InstructX2X, an interpretable local editing model that leverages region-specific editing and a new expert-verified dataset (MIMIC-EDIT-INSTRUCTION) to generate high-quality counterfactual medical images while preventing unintended demographic changes and providing visual explanations for the editing process.
The paper introduces Fact-Flow, a novel framework that enhances the factual accuracy of MLLM-based medical report generation by decoupling visual fact identification from text generation and utilizing an LLM-bootstrapped pipeline to create labeled training data without manual annotation.
This paper proposes Taxonomy-Aware Representation Alignment (TARA), a method that enhances Large Multimodal Models' hierarchical visual recognition capabilities for both known and novel categories by aligning their visual representations with biology foundation models and ground-truth labels to enforce taxonomic consistency.
This paper proposes TAP-SLF, a parameter-efficient framework that combines task-aware soft prompts and selective fine-tuning of top encoder layers to effectively adapt Vision Foundation Models for multi-task ultrasound image analysis while minimizing overfitting and computational costs.
This paper introduces ICLA, an internal self-correction mechanism that leverages a diagonal cross-layer attention mechanism to enable Large Vision-Language Models to refine their own hidden states and mitigate hallucinations without external signals, demonstrating consistent improvements across benchmarks with minimal additional parameters.
Mamba-CAD is a self-supervised generative modeling framework that leverages a Mamba-based encoder-decoder architecture and a new large-scale dataset to effectively generate complex, long-sequence parametric CAD models for industrial applications.
SesaHand is a novel framework that enhances 3D hand reconstruction by generating diverse, high-quality synthetic hand images through a controllable generation pipeline that ensures semantic alignment via Chain-of-Thought reasoning and structural alignment via hierarchical fusion and attention mechanisms.
This paper proposes an improved adversarial diffusion compression method that distills a heavy 3D diffusion Transformer into a lightweight 2D-based model with 1D temporal convolutions and a dual-head adversarial scheme, achieving a 95% reduction in parameters and 8 speedup while effectively balancing spatial detail and temporal consistency for real-world video super-resolution.
The paper introduces LSS-LTCNet, an efficient and explainable framework that combines Local Self-Similarity texture priors with continuous-time neural dynamics to achieve state-of-the-art foot ulcer segmentation and boundary precision on the MICCAI FUSeg dataset.
This paper introduces ReMoT, a unified training paradigm that combines a rule-based framework for generating a large-scale motion-contrast dataset with Group Relative Policy Optimization to significantly enhance VLMs' spatio-temporal consistency and reasoning capabilities, achieving state-of-the-art performance on both new and standard benchmarks.
This paper introduces OPGAgent, a multi-tool agentic system that enhances the accuracy and audibility of dental panoramic X-ray interpretation by coordinating specialized perception modules through a hierarchical evidence gathering process and a consensus mechanism, while also proposing the OPG-Bench benchmark for comprehensive evaluation beyond standard VQA metrics.
DreamWorld introduces a unified framework that integrates complementary world knowledge into video generation through a Joint World Modeling Paradigm, employing Consistent Constraint Annealing and Multi-Source Inner-Guidance to overcome visual instability and achieve superior temporal, spatial, and semantic consistency compared to existing models.
This paper presents a hardware-algorithm co-designed HDR imaging system that integrates an asymmetric event-SVE camera with a novel two-stage alignment framework and a cross-modal reconstruction network to achieve superior highlight recovery and edge fidelity in extreme illumination conditions.
This paper introduces FEWTRANS, a comprehensive benchmark and the Hyperparameter Ensemble (HPE) evaluation protocol to rigorously assess few-shot transfer learning, revealing that pre-trained model selection and full-parameter fine-tuning often outperform sophisticated adaptation methods due to their ability to make distributed micro-adjustments without overfitting.
The paper introduces U-VLM, a hierarchical vision-language model that combines progressive multi-stage training with multi-layer visual feature injection to achieve state-of-the-art radiology report generation on 3D medical imaging, demonstrating that specialized encoder pretraining can outperform massive language models.
This paper presents a probabilistic framework to quantify the number of adversarial ballots required to flip a U.S. election and empirically demonstrates through 144,000 physical print-and-scan experiments that the most effective adversarial attacks in the physical voting domain differ significantly from those in the digital domain.