AGI-survey

AGI-survey

人工通用智能研究前沿及未来发展路线图概览

AGI-survey项目系统梳理了人工通用智能(AGI)研究的前沿进展。项目覆盖AGI内部机制、接口设计、系统实现、对齐问题及发展路线等核心领域,汇总分析了大量相关论文。内容涉及AGI的感知、推理、记忆能力,及其与数字世界、物理世界和其他智能体的交互。此外,项目还探讨了AGI的评估方法和伦理考量,为AGI的发展提供全面参考。

AGI人工智能大语言模型多模态推理Github开源项目

Awesome AGI Survey<br><sub>Must-read papers on Artificial General Intelligence</sub>

Arxiv Paper Workshop Link License: MIT

<p align="center"> <img src="https://github.com/JiaxuanYou/LLM-AGI/blob/main/assets/fig/abstract.png" alt="Abstract Image"> </p>

🔔 News

🔥 Our project is an ongoing, open initiative that will evolve in parallel with advancements in AGI. We plan to add more work soon, and we highly welcome pull requests!

BibTex citation if you find our work/resources useful:

@article{feng2024far, title={How Far Are We From AGI}, author={Feng, Tao and Jin, Chuanyang and Liu, Jingyu and Zhu, Kunlun and Tu, Haoqin and Cheng, Zirui and Lin, Guanyu and You, Jiaxuan}, journal={arXiv preprint arXiv:2405.10313}, year={2024} }

📜Content

<p align="center"> <figure> <img src="https://github.com/JiaxuanYou/LLM-AGI/blob/main/assets/fig/intro.jpg" alt="intro"> </figure> </p>

-> The framework design of our paper. <-

1. Introduction

<p align="center"> <figure> <img src="https://github.com/JiaxuanYou/LLM-AGI/blob/main/assets/fig/f2-1.png" width="600"> </figure> </p>

-> Proportion of Human Activities Surpassed by AI. <-

2. AGI Internal: Unveiling the Mind of AGI

<p align="center"> <img src="https://github.com/JiaxuanYou/LLM-AGI/blob/main/assets/fig/2a-1.png" width="600"> </p> <p align="center"> <img src="https://github.com/JiaxuanYou/LLM-AGI/blob/main/assets/fig/2b-1.png" width="600"> </p>

2.1 AI Perception

  1. Flamingo: a Visual Language Model for Few-Shot Learning. Jean-Baptiste Alayrac et al. NeurIPS 2022. [paper]
  2. BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models. Junnan Li et al. ICML 2023. [paper]
  3. SPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models. Ziyi Lin et al. EMNLP 2023. [paper]
  4. Visual Instruction Tuning. Haotian Liu et al. NeurIPS 2023. [paper]
  5. GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction. Rui Yang et al. NeurIPS 2023. [paper]
  6. Otter: A Multi-Modal Model with In-Context Instruction Tuning. Bo Li et al. arXiv 2023. [paper]
  7. VideoChat: Chat-Centric Video Understanding. KunChang Li et al. arXiv 2023. [paper]
  8. mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality. Qinghao Ye et al. arXiv 2023. [paper]
  9. A Survey on Multimodal Large Language Models. Shukang Yin et al. arXiv 2023. [paper]
  10. PandaGPT: One Model To Instruction-Follow Them All. Yixuan Su et al. arXiv 2023. [paper]
  11. LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention. Renrui Zhang et al. arXiv 2023. [paper]
  12. Gemini: A Family of Highly Capable Multimodal Models. Rohan Anil et al. arXiv 2023. [paper]
  13. Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic. Keqin Chen et al. arXiv 2023. [paper]
  14. ImageBind: One Embedding Space To Bind Them All. Rohit Girdhar et al. CVPR 2023. [paper]
  15. MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices. Xiangxiang Chu et al. arXiv 2023. [paper]
  16. What Makes for Good Visual Tokenizers for Large Language Models?. Guangzhi Wang et al. arXiv 2023. [paper]
  17. MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models. Deyao Zhu et al. ICLR 2024. [paper]
  18. LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment. Bin Zhu et al. ICLR 2024. [paper]

2.2 AI Reasoning

  1. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. Jason Wei et al. NeurIPS 2022. [paper]
  2. Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs. Maarten Sap et al. EMNLP 2022. [paper]
  3. Inner Monologue: Embodied Reasoning through Planning with Language Models. Wenlong Huang et al. CoRL 2022. [paper]
  4. Survey of Hallucination in Natural Language Generation. Ziwei Ji et al. ACM Computing Surveys 2022. [paper]
  5. ReAct: Synergizing Reasoning and Acting in Language Models. Shunyu Yao et al. ICLR 2023. [paper]
  6. Decomposed Prompting: A Modular Approach for Solving Complex Tasks. Tushar Khot et al. ICLR 2023. [paper]
  7. Complexity-Based Prompting for Multi-Step Reasoning. Yao Fu et al. ICLR 2023. [paper]
  8. Least-to-Most Prompting Enables Complex Reasoning in Large Language Models. Denny Zhou et al. ICLR 2023. [paper]
  9. Towards Reasoning in Large Language Models: A Survey. Jie Huang et al. ACL Findings 2023. [paper]
  10. ProgPrompt: Generating Situated Robot Task Plans using Large Language Models. Ishika Singh et al. ICRA 2023. [paper]
  11. Reasoning with Language Model is Planning with World Model. Shibo Hao et al. EMNLP 2023. [paper]
  12. Evaluating Object Hallucination in Large Vision-Language Models. Yifan Li et al. EMNLP 2023. [paper]
  13. Tree of Thoughts: Deliberate Problem Solving with Large Language Models. Shunyu Yao et al. NeurIPS 2023. [paper]
  14. Self-Refine: Iterative Refinement with Self-Feedback. Aman Madaan et al. NeurIPS 2023. [paper]
  15. Reflexion: Language Agents with Verbal Reinforcement Learning. Noah Shinn et al. NeurIPS 2023. [paper]
  16. Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents. Zihao Wang et al. NeurIPS 2023. [paper]
  17. LLM+P: Empowering Large Language Models with Optimal Planning Proficiency. Bo Liu et al. arXiv 2023. [paper]
  18. Language Models, Agent Models, and World Models: The LAW for Machine Reasoning and Planning. Zhiting Hu et al. arXiv 2023. [paper]
  19. MMToM-QA: Multimodal Theory of Mind Question Answering. Chuanyang Jin et al. arXiv 2024. [paper]
  20. Graph of Thoughts: Solving Elaborate Problems with Large Language Models. Maciej Besta et al. AAAI 2024. [paper]
  21. Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Perfect Reasoners. Qihuang Zhong et al. arXiv 2024. [paper] pending

2.3 AI Memory

  1. Dense Passage Retrieval for Open-Domain Question Answering. Vladimir Karpukhin et al. EMNLP 2020. [paper]
  2. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. Patrick Lewis et al. NeurIPS 2020. [paper]
  3. REALM: Retrieval-Augmented Language Model Pre-Training. Kelvin Guu et al. ICML 2020. [paper]
  4. Retrieval Augmentation Reduces Hallucination in Conversation. Kurt Shuster et al. EMNLP Findings 2021. [paper]
  5. Improving Language Models by Retrieving from Trillions of Tokens. Sebastian Borgeaud et al. ICML 2022. [paper]
  6. Generative Agents: Interactive Simulacra of Human Behavior. Joon Sung Park et al. UIST 2023. [paper]
  7. Cognitive Architectures for Language Agents. Theodore R. Sumers et al. TMLR 2024. [paper]
  8. Voyager: An Open-Ended Embodied Agent with Large Language Models. Guanzhi Wang et al. arXiv 2023. [paper]
  9. **A Survey on the Memory Mechanism of Large Language Model based

编辑推荐精选

openai-agents-python

openai-agents-python

OpenAI Agents SDK,助力开发者便捷使用 OpenAI 相关功能。

openai-agents-python 是 OpenAI 推出的一款强大 Python SDK,它为开发者提供了与 OpenAI 模型交互的高效工具,支持工具调用、结果处理、追踪等功能,涵盖多种应用场景,如研究助手、财务研究等,能显著提升开发效率,让开发者更轻松地利用 OpenAI 的技术优势。

Hunyuan3D-2

Hunyuan3D-2

高分辨率纹理 3D 资产生成

Hunyuan3D-2 是腾讯开发的用于 3D 资产生成的强大工具,支持从文本描述、单张图片或多视角图片生成 3D 模型,具备快速形状生成能力,可生成带纹理的高质量 3D 模型,适用于多个领域,为 3D 创作提供了高效解决方案。

3FS

3FS

一个具备存储、管理和客户端操作等多种功能的分布式文件系统相关项目。

3FS 是一个功能强大的分布式文件系统项目,涵盖了存储引擎、元数据管理、客户端工具等多个模块。它支持多种文件操作,如创建文件和目录、设置布局等,同时具备高效的事件循环、节点选择和协程池管理等特性。适用于需要大规模数据存储和管理的场景,能够提高系统的性能和可靠性,是分布式存储领域的优质解决方案。

TRELLIS

TRELLIS

用于可扩展和多功能 3D 生成的结构化 3D 潜在表示

TRELLIS 是一个专注于 3D 生成的项目,它利用结构化 3D 潜在表示技术,实现了可扩展且多功能的 3D 生成。项目提供了多种 3D 生成的方法和工具,包括文本到 3D、图像到 3D 等,并且支持多种输出格式,如 3D 高斯、辐射场和网格等。通过 TRELLIS,用户可以根据文本描述或图像输入快速生成高质量的 3D 资产,适用于游戏开发、动画制作、虚拟现实等多个领域。

ai-agents-for-beginners

ai-agents-for-beginners

10 节课教你开启构建 AI 代理所需的一切知识

AI Agents for Beginners 是一个专为初学者打造的课程项目,提供 10 节课程,涵盖构建 AI 代理的必备知识,支持多种语言,包含规划设计、工具使用、多代理等丰富内容,助您快速入门 AI 代理领域。

AEE

AEE

AI Excel全自动制表工具

AEE 在线 AI 全自动 Excel 编辑器,提供智能录入、自动公式、数据整理、图表生成等功能,高效处理 Excel 任务,提升办公效率。支持自动高亮数据、批量计算、不规则数据录入,适用于企业、教育、金融等多场景。

UI-TARS-desktop

UI-TARS-desktop

基于 UI-TARS 视觉语言模型的桌面应用,可通过自然语言控制计算机进行多模态操作。

UI-TARS-desktop 是一款功能强大的桌面应用,基于 UI-TARS(视觉语言模型)构建。它具备自然语言控制、截图与视觉识别、精确的鼠标键盘控制等功能,支持跨平台使用(Windows/MacOS),能提供实时反馈和状态显示,且数据完全本地处理,保障隐私安全。该应用集成了多种大语言模型和搜索方式,还可进行文件系统操作。适用于需要智能交互和自动化任务的场景,如信息检索、文件管理等。其提供了详细的文档,包括快速启动、部署、贡献指南和 SDK 使用说明等,方便开发者使用和扩展。

Wan2.1

Wan2.1

开源且先进的大规模视频生成模型项目

Wan2.1 是一个开源且先进的大规模视频生成模型项目,支持文本到图像、文本到视频、图像到视频等多种生成任务。它具备丰富的配置选项,可调整分辨率、扩散步数等参数,还能对提示词进行增强。使用了多种先进技术和工具,在视频和图像生成领域具有广泛应用前景,适合研究人员和开发者使用。

爱图表

爱图表

全流程 AI 驱动的数据可视化工具,助力用户轻松创作高颜值图表

爱图表(aitubiao.com)就是AI图表,是由镝数科技推出的一款创新型智能数据可视化平台,专注于为用户提供便捷的图表生成、数据分析和报告撰写服务。爱图表是中国首个在图表场景接入DeepSeek的产品。通过接入前沿的DeepSeek系列AI模型,爱图表结合强大的数据处理能力与智能化功能,致力于帮助职场人士高效处理和表达数据,提升工作效率和报告质量。

Qwen2.5-VL

Qwen2.5-VL

一款强大的视觉语言模型,支持图像和视频输入

Qwen2.5-VL 是一款强大的视觉语言模型,支持图像和视频输入,可用于多种场景,如商品特点总结、图像文字识别等。项目提供了 OpenAI API 服务、Web UI 示例等部署方式,还包含了视觉处理工具,有助于开发者快速集成和使用,提升工作效率。

下拉加载更多