AI研究論文デイリー

Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs

MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators

SwapAnything: Enabling Arbitrary Object Swapping in Personalized Visual Editing

ByteEdit: Boost, Comply and Accelerate Generative Image Editing

UniFL: Improve Stable Diffusion via Unified Feedback Learning

SpatialTracker: Tracking Any 2D Pixels in 3D Space

BeyondScene: Higher-Resolution Human-Centric Scene Generation With Pretrained Diffusion

MA-LMM: Memory-Augmented Large Multimodal Model for Long-Term Video Understanding

PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations

YaART: Yet Another ART Rendering Technology

MoMA: Multimodal LLM Adapter for Fast Personalized Image Generation

Aligning Diffusion Models by Optimizing Human Utility

Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models

DATENeRF: Depth-Aware Text-based Editing of NeRFs

Koala: Key frame-conditioned long video-LLM