Ежедневные Исследовательские Статьи по ИИ

VidTwin: Video VAE with Decoupled Structure and Dynamics

A Silver Bullet or a Compromise for Full Attention? A Comprehensive Study of Gist Token-based Context Compression

MMFactory: A Universal Solution Search Engine for Vision-Language Tasks

Molar: Multimodal LLMs with Collaborative Filtering Alignment for Enhanced Sequential Recommendation

YuLan-Mini: An Open Data-efficient Language Model