AI 연구 논문 데일리

LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens

YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information

Aria Everyday Activities Dataset

SDXL-Lightning: Progressive Adversarial Diffusion Distillation

Snap Video: Scaled Spatiotemporal Transformers for Text-to-Video Synthesis

User-LLM: Efficient LLM Contextualization with User Embeddings

In deep reinforcement learning, a pruned network is a good network

Coercing LLMs to do and reveal (almost) anything

Music Style Transfer with Time-Varying Inversion of Diffusion Models

BBA: Bi-Modal Behavioral Alignment for Reasoning with Large Vision-Language Models

ToDo: Token Downsampling for Efficient Generation of High-Resolution Images

D-Flow: Differentiating through Flows for Controlled Generation

Ouroboros: Speculative Decoding with Large Model Enhanced Drafting