[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-80746":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":28,"readmeContent":29,"aiSummary":30,"trendingCount":15,"starSnapshotCount":15,"syncStatus":14,"lastSyncTime":31,"discoverSource":32},80746,"paper-orchestra","google-research\u002Fpaper-orchestra","google-research","PaperOrchestra is a multi-agent framework for automated AI research paper writing.","https:\u002F\u002Fyiwen-song.github.io\u002Fpaper_orchestra\u002F",null,"Python",55,8,2,0,5,9,14,15,2.86,"Apache License 2.0",false,"main",true,[26,27],"ai-research-assistant","multi-agent-systems","2026-06-12 02:04:06","# \u003Cdiv align=\"center\">🎻 PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing\u003C\u002Fdiv>\n\u003Cdiv align=\"center\">Yiwen Song\u003Csup>1\u003C\u002Fsup>, Yale Song\u003Csup>1\u003C\u002Fsup>, Tomas Pfister\u003Csup>1\u003C\u002Fsup>, and Jinsung Yoon\u003Csup>1\u003C\u002Fsup>\u003C\u002Fdiv>\n\u003Cdiv align=\"center\">\u003Csup>1\u003C\u002Fsup>Google Cloud AI Research\u003C\u002Fdiv>\n\u003Cbr>\u003Cbr>\n\nPaperOrchestra is a multi-agent framework for automated AI research paper writing. It flexibly transforms unconstrained pre-writing materials (ideas and experimental logs) into submission-ready LaTeX manuscripts, including comprehensive literature synthesis and generated visuals, such as plots and conceptual diagrams.\n\nActing like an orchestrated team of specialized agents, it handles outline generation, literature review, section writing, content refinement, and plotting to close the gap between raw research materials and submission ready papers.\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"assets\u002Foverview.png\" alt=\"PaperOrchestra Overview\" width=\"90%\"\u002F>\n\u003C\u002Fdiv>\n\n## Key Features\n\n- **Multi-Agent Pipeline**: Orchestrates specialized agents (Outline, Literature Review, Section Writing, Content Refinement, Plotting) for end-to-end paper generation.\n- **Comprehensive Literature Synthesis**: Automatically searches, prioritizes, and integrates relevant literature into the manuscript.\n- **Visuals Generation**: Generates plots and supports the inclusion of conceptual diagrams.\n- **LaTeX Automation**: Produces submission-ready LaTeX manuscripts adhering to target conference templates.\n\n## Installation\n\n1.  **Environment Setup**: The project relies on a Conda environment named `paper_orchestra`. You can set up a similar environment with Python 3.11 and install the required dependencies.\n    ```bash\n    conda create -n paper_orchestra python=3.11\n    conda activate paper_orchestra\n    pip install -r requirements.txt\n    ```\n\n2.  **Environment Variables**: Create a `.env` file in the root of the project and add your API keys:\n    ```env\n    # Add keys if you have them, otherwise leave them empty\n    # You need to specify related keys if you want to use the model\n    # For example, if you want to use Gemini models, you need to specify either VERTEX_AI_PROJECT and VERTEX_AI_LOCATION, or GEMINI_API_KEY\n    # If you want to use OpenAI models, you need to specify OPENAI_API_KEY\n    \n    SEMANTIC_SCHOLAR_API_KEY=your_semantic_scholar_api_key\n    OPENAI_API_KEY=your_openai_api_key\n    VERTEX_AI_PROJECT=your_vertex_ai_project\n    VERTEX_AI_LOCATION=your_vertex_ai_location\n    GEMINI_API_KEY=your_gemini_api_key\n\n    # if you want to run demo and send emails to yourself at the end, set these 2 fields as well\n    SMTP_EMAIL=your_email\n    SMTP_PASSWORD=your_password\n    ```\n\n## Dataset\n\nThe dataset used for paper generation is **not included** in this code repository. It will be released separately at a later date.\nOnce available, you should download the dataset and place it under the `datasets\u002F` directory.\n\n## Usage\n\nYou can run the paper generation pipeline using the provided bash script or directly via the Python CLI.\n\n### Using the Bash Script\n```bash\n.\u002Fpaper_writing_cli.sh\n```\n\n### Using the Python CLI directly\n```bash\npython paper_writing_cli.py \\\n  --raw_materials_dir \u002Fpath_to_your_raw_materials\u002F \\\n  --latex_template_dir \u002Fpath_to_your_latex_template\u002F \\\n  --output_dir \u002Fpath_to_your_output_folder\u002F \\\n  --use_plotting true\n```\n\n### Key Arguments\n*   `--raw_materials_dir`: (Required) Directory containing the raw materials (e.g., idea and experimental log).\n*   `--latex_template_dir`: (Required) Directory containing the LaTeX template (e.g., `templates\u002Fcvpr2025`).\n*   `--output_dir`: Directory to output the generated paper. If not specified, an automatic folder with a timestamp is created.\n*   `--use_plotting`: Enable the plotting agent workflow to generate figures from code (Default: `False`).\n*   `--writer_model_name`: LLM for writer and literature agents.\n*   `--reflection_model_name`: LLM for reflection agents.\n\n### Interactive Demo\n\nWe provide an interactive Streamlit demo for the paper generation pipeline. For instructions on how to run and use it, please refer to the [README.md](frontend\u002FREADME.md) in the `frontend` directory.\n\n## Project Structure\n\n*   `autoraters\u002F`: Contains scripts for automated evaluation.\n*   `methods\u002F`: Core logic for agents and pipeline execution.\n*   `templates\u002F`: LaTeX templates for target conferences, you can add your own template by creating a new subfolder under this directory.\n*   `utils\u002F`: Utility functions for API calls, PDF handling, etc.\n\n## Citation\n\nIf you find this repo or our paper helpful, please cite it as follows:\n\n```bibtex\n@article{song2026paperorchestra,\n  title={PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing},\n  author={Song, Yiwen and Song, Yale and Pfister, Tomas and Yoon, Jinsung},\n  journal={arXiv preprint arXiv:2604.05018},\n  year={2026}\n}\n```\n\n# Disclaimer\n\nThis is not an officially supported Google product.\n","PaperOrchestra 是一个用于自动生成AI研究论文的多智能体框架。它能够将初步的研究材料（如想法和实验记录）灵活地转化为符合提交要求的LaTeX文档，包括全面的文献综述和生成的可视化图表。该项目的核心功能在于其多智能体流水线设计，通过协调专门化的代理（负责提纲生成、文献回顾、章节撰写、内容优化及绘图等任务），实现从原始研究资料到准备提交的论文的全流程自动化处理。此外，它还支持自动检索并整合相关文献，并能根据需要生成或包含概念性图表。PaperOrchestra 适用于希望提高科研写作效率的研究人员以及团队，特别是在面对时间紧迫或需要快速整理大量信息的情况下尤为有用。","2026-06-11 04:01:53","CREATED_QUERY"]