AI Papers

A New Era of Scene Image Editing with Enhanced Control and Precision Unified 2D to 3D Editing: 3DitScene introduces a seamless framework for editing scenes from...

A New Approach to Reducing Memory Consumption in Training Large Language Models VeLoRA introduces rank-1 sub-token projections to significantly reduce memory requirements during model training. The...

Enhancing 3D Models with Structural Detail from Single-view Images Innovative Multiview Diffusion Technique: Uses diffusion models to create multiview images for accurate 3D reconstruction. Part-aware Segmentation:...

Revolutionizing Human Video Generation for Virtual Reality and Animation Innovative 4D Transformer Architecture: Efficient modeling of spatio-temporal correlations across viewpoints and time. Precise Conditioning Mechanism: Utilizes...

Transforming Video Generation for Enhanced AI Interactivity Scalable Autoregressive Transformer: iVideoGPT integrates multimodal signals into a sequence of tokens for interactive AI experiences. Compressive Tokenization Technique:...

Leveraging Multifaceted Rationales for Superior Performance Unified Transformer Model: Meteor leverages the Mamba architecture to efficiently embed multifaceted rationales. Enhanced Performance: Significant improvements in vision-language tasks...

Exploring a Lightweight Approach to Bridging Visual and Audio Generation Unified Transformer Model: Visual Echoes uses a simple generative transformer for both audio-visual generation and...

New AI Technique Promises Better Cross-Domain Generalization in Image Matching Foundation Model Guidance: OmniGlue uses a vision foundation model to improve feature matching across different...

IDEA Research Introduces High-Performance and Efficient Models for Enhanced Object Detection Two Advanced Models: Grounding DINO 1.5 Pro and Grounding DINO 1.5 Edge offer high-performance...

New Method Generates 3D Scenes Quickly and Efficiently from Minimal Inputs Efficient 3D Generation: CAT3D uses multi-view diffusion models to generate consistent 3D scenes from...

Huawei's Framework Offers New Insights Beyond Traditional Scaling Laws Associative Memory Modeling: Transformers are modeled using associative memories, explaining the attention mechanism through Hopfield networks. Energy...

New Framework Allows Users to Control and Edit 3D Models with Ease Interactive Generation Workflow: Coin3D enables users to control 3D generation using coarse geometry...