AI 연구 논문 데일리

Simple and Scalable Strategies to Continually Pre-train Large Language Models

Gemma: Open Models Based on Gemini Research and Technology

VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis

SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents

On the Societal Impact of Open Foundation Models

Scaling Up Dynamic Human-Scene Interaction Modeling

Language models scale reliably with over-training and on downstream tasks

Follow-Your-Click: Open-domain Regional Image Animation via Short Prompts

GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Splatting