Awesome World Models / arXiv

Emu3.5: Native Multimodal Models are World Learners

作者：作者信息待提取 · 来源：General Approaches to World Models · 发布日期：2025 · 分类：2. Building World Models from 2D Vision Priors / To what extent does Vision Intelligence exist in Video Generation Models

二维视觉先验视觉语言桥接视频生成
新标签打开 AlphaXiv新标签打开 Project
Emu3.5: Native Multimodal Models are World Learners figure
AlphaXiv 中文概览（可滚动查看）