AI研究論文デイリー

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction

Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models

ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline

InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation

On the Scalability of Diffusion-based Text-to-Image Generation

Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models

Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition