{"id":26896961,"url":"https://github.com/JoeLeelyf/customize-arxiv-daily","last_synced_at":"2025-04-01T04:02:33.200Z","repository":{"id":278560638,"uuid":"936020115","full_name":"JoeLeelyf/customize-arxiv-daily","owner":"JoeLeelyf","description":"Customize your arXiv recommendation every day.","archived":false,"fork":false,"pushed_at":"2025-03-25T03:54:30.000Z","size":2384,"stargazers_count":94,"open_issues_count":0,"forks_count":13,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-25T04:28:29.392Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/JoeLeelyf.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-02-20T11:58:12.000Z","updated_at":"2025-03-25T03:54:33.000Z","dependencies_parsed_at":"2025-02-20T13:25:09.797Z","dependency_job_id":"c031994d-c1dc-4969-9a21-8dfce376f2c1","html_url":"https://github.com/JoeLeelyf/customize-arxiv-daily","commit_stats":null,"previous_names":["joeleelyf/customize-arxiv-daily"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JoeLeelyf%2Fcustomize-arxiv-daily","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JoeLeelyf%2Fcustomize-arxiv-daily/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JoeLeelyf%2Fcustomize-arxiv-daily/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/JoeLeelyf%2Fcustomize-arxiv-daily/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/JoeLeelyf","download_url":"https://codeload.github.com/JoeLeelyf/customize-arxiv-daily/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246580468,"owners_count":20800111,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-04-01T04:02:25.395Z","updated_at":"2025-04-01T04:02:33.188Z","avatar_url":"https://github.com/JoeLeelyf.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"\u003ch3 align=\"center\"\u003eCustomize-arXiv-Daily\u003c/h3\u003e\n\n---\n\n\u003cp align=\"center\"\u003e Recommend new arxiv papers of your interest daily according to your customized description.\n    \u003cbr\u003e \n\u003c/p\u003e\n\n\u003e [!NOTE]\n\u003e This repo borrow the idea and some functions from [zotero-arxiv-daily](https://github.com/TideDra/zotero-arxiv-daily). Thanks for their great work!😊\n\n## 🧐 Why I create this project \u003ca name = \"about\"\u003e\u003c/a\u003e\n\n- During the use of [zotero-arxiv-daily](https://github.com/TideDra/zotero-arxiv-daily), I often find that the recommendation process didn't run in the way that I want. Since my study area has shifted, my Zotero include some papers that I'm not interested in anymore.\n- For those who **do not use zotero as PDF reader**, get customized arxiv recommendation is still needed.\n- For those that want to **set their own prompt** to guide LLM during paper selection and recommendation.\n- For those who want to leverage convenient local deployment tools, like **Ollama**\n\n## ✨ Key Features Compared with [zotero-arxiv-daily](https://github.com/TideDra/zotero-arxiv-daily)\n\n- Fully customized LLM prompt to guide your paper recommendation process.\n- Ready-to-use leverage of recent models, include DeepSeek-R1/V3/...\n- Use Ollama to accelerate local deployment.\n- Save your arXiv recommendation history.\n- Summarize and recommend papers to you at the start of the email.\n- Support multiple workers to speed up the recommendation process.\n\n## 📷 Screenshot\n\n![screenshot](./assets/screenshot.png)\n\n## 🚀 Usage\n\n### Quick Start\n\n1. Run `git clone https://github.com/JoeLeelyf/customize-arxiv-daily.git`\n2. Run `pip install -r requirements.txt` to install necessary packages.\n3. Get your STMP server. Common STMP service provider includes [QQ mail box](https://service.mail.qq.com/detail/0/427)\n4. Describe the research fields you're interested in, and the fields you're not. Edit the `description.txt`. For, example:\n\n```txt\nI am working on the research area of computer vision.\nSpecifically, I am interested in the following fieds:\n1. Object detection\n2. AIGC (AI Generated Content)\n3. Multimodal Large Language Models\n\nI'm not interested in the following fields:\n1. 3D Vision\n2. Robotics\n3. Low-level Vision\n```\n\n5. Configure your own `arXiv catergories`, `api_key` and `models`. The repo currently support any api in OpenAI api form, and various models in Ollama. Meaning of different parameters:\n   - `--categories`: arXiv categories that you are interested in, like `cs.CV` `cs.AI`\n   - `--sender`: E-mail address that provide SMTP service, like, `123456@qq.com`\n   - `--receiver`: The e-mails address that you want to receive your notice, like, `my_gmail@gmail.com`\n   - `--save`: store_true, whether to save the arXiv results to local markdown files.\n\n- `main_gpt.sh`: Visit [OpenAI](https://openai.com/) to get a OpenAI api_key invole GPT models.\n\n```bash\npython main.py --categories cs.CV cs.AI \\\n    --provider OpenAI --model gpt-4o \\\n    --base_url https://api.openai.com/v1 --api_key * \\\n    --smtp_server smtp.qq.com --smtp_port 465 \\\n    --sender * --receiver * \\\n    --sender_password * \\\n    --num_workers 16 \\\n    --title \"Daily arXiv\" \\\n    --temperature 0.7 \\\n    --save\n```\n\n- `main_silicon_flow.sh`: Visit [SiliconFlow](https://siliconflow.cn/zh-cn/) to get a SiliconFlow api_key for DeepSeek models and more.\n\n```bash\npython main.py --categories cs.CV cs.AI\\\n    --provider SiliconFlow --model deepseek-ai/DeepSeek-R1-Distill-Llama-70B \\\n    --base_url https://api.siliconflow.cn/v1 --api_key * \\\n    --smtp_server smtp.qq.com --smtp_port 465 \\\n    --sender * --receiver * \\\n    --sender_password * \\\n    --num_workers 16 \\\n    --title \"Daily arXiv\" \\\n    --temperature 0.7 \\\n    --save\n```\n\n- `main_ollama.sh`: Visit [Ollama](https://ollama.com/) for local deployment.\n\n```bash\npython main.py --categories cs.CV cs.AI\\\n    --provider Ollama --model deepseek-r1:7b \\\n    --smtp_server smtp.qq.com --smtp_port 465 \\\n    --sender * --receiver * \\\n    --sender_password * \\\n    --num_workers 16 \\\n    --title \"Daily arXiv\" \\\n    --temperature 0.7 \\\n    --save\n```\n\n6. Choose to run one of the following command in your CLI.\n\n```\nbash main_gpt.sh\nbash main_silicon_flow.sh\nbash main_ollama.sh\n```\n\n7. \\* **Run automatically everyday.** For example, use `crontab` on Linux to run the script periodically:\n\n- Edit `crontab` file:\n\n```bash\ncrontab -e\n```\n\n- Add following line and run the script in 5:00 AM everyday (anytime you want):\n\n```txt\n* 5 * * * /path/to/customize-arxiv-daily/main_gpt.sh\n```\n\n8. \\* **Adjust and customize your LLM prompt.** Edit `def get_response(self, title, abstract):` func in `arxiv_daily.py`\n\n## Results\n\n### Running process in your CLI\n\n![CLI](./assets/cli.png)\n\n### Markdown saved\n\n![Markdown](./assets/markdown.png)\n\n### E-mail received\n\n![Screenshot](./assets/screenshot.png)\n\n## 📖 How it works\n\n- `util/request.py` crawls the arXiv web page given your provided arXiv categories.\n- `arxiv_daily` will call LLM api to summarize every paper and get the relevance score.\n- `util/construct_email.py` construct the content of the email in HTML form and send it using SMTP service.\n\n## 📌 Limitations\n\n- The recommendation process of LLM is unstable and the relevance score provided by different LLMs varies a lot.\n\n## ❤️ Acknowledgement\n\n- [zotero-arxiv-daily](https://github.com/TideDra/zotero-arxiv-daily)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FJoeLeelyf%2Fcustomize-arxiv-daily","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FJoeLeelyf%2Fcustomize-arxiv-daily","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FJoeLeelyf%2Fcustomize-arxiv-daily/lists"}