-
Scaling Laws in Patchification: An Image Is Worth 50,176 Tokens And More
Paper • 2502.03738 • Published • 10 -
Better Embeddings with Coupled Adam
Paper • 2502.08441 • Published • 1 -
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment
Paper • 2502.16894 • Published • 21
Oğuzhan Ercan
oguzhanercan
AI & ML interests
Computer Vision, Generative Vision, first trajectory bender
Recent Activity
updated
a collection
1 day ago
Training Theory
updated
a collection
1 day ago
Image-Video General Tasks
updated
a collection
1 day ago
Diffusion/Flow Model Optimization
Organizations
None yet
Collections
24
-
QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation
Paper • 2502.05178 • Published • 10 -
Scaling Text-Rich Image Understanding via Code-Guided Synthetic Multimodal Data Generation
Paper • 2502.14846 • Published • 13 -
SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
Paper • 2502.14786 • Published • 115
models
None public yet