简介
Agentic world modeling 关注 world model 与 agent 行为的结合;该目录有助于你梳理关键论文与评测方法。
最适合: 在设计 world-model 能力与评测体系的研究者
适配: 任意系统;Markdown 目录;完整内容需进入各条目源链接阅读
配置时间: 5–15 分钟
关键事实(已验证)
- 目录仓库为 MIT 许可证(GitHub API 已验证)。
- 建议先明确“world model”定义(状态/动力学/记忆/规划),再选择对应评测与数据集。
- GitHub:209 stars · 5 forks;最近更新 2026-05-11(GitHub API 验证)。
正文
建议用“定义 → 评测 → 实验”来沉淀:
- 选 3 篇基础论文,写一页纸的“共同定义”文档。
- 收集评测方法,明确对你的 agent 最重要的指标(准确率、规划深度、泛化等)。
- 把阅读转成实验:先做最小 baseline,再逐步加复杂度。
README 原文节选(verbatim)
Awesome Agentic World Modeling
This repository accompanies the Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond, providing a taxonomy-aligned bibliography of 400+ cited works and 100+ representative systems. Papers are grouped by taxonomy section and listed in reverse chronological order within each subsection to support literature navigation, comparison, and ongoing updates. Released under the MIT License. Check out our poster here.
[!TIP] 👋 Welcome to join the discussion on
or
, share your work in progress, and help us grow the agentic world modeling community together.
We also welcome collaborations on distilled surveys, follow-up research, and related projects across the physical, digital, social, and scientific world modeling.
[!NOTE] 📚 If you find this resource useful, please cite and
the repo:
@article{chu2026agenticworldmodelingfoundations, title = {Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond},
FAQ
world modeling 只适用于机器人吗? 答:不一定:相关工作也覆盖游戏与通用 agent 规划场景。
先读什么更好? 答:建议先读定义/综述,再进入评测与 benchmark。
怎么让笔记更有用? 答:记录假设、指标与“你实际做过实验验证”的短名单。