{"id":42757876,"url":"https://github.com/sciknoworg/llm-atlas","last_synced_at":"2026-01-29T20:17:02.018Z","repository":{"id":334792035,"uuid":"1135585112","full_name":"sciknoworg/llm-atlas","owner":"sciknoworg","description":null,"archived":false,"fork":false,"pushed_at":"2026-01-26T22:58:49.000Z","size":226,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-01-27T08:47:58.546Z","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/sciknoworg.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,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-01-16T09:58:31.000Z","updated_at":"2026-01-26T22:58:53.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/sciknoworg/llm-atlas","commit_stats":null,"previous_names":["sciknoworg/llm-atlas"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/sciknoworg/llm-atlas","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sciknoworg%2Fllm-atlas","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sciknoworg%2Fllm-atlas/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sciknoworg%2Fllm-atlas/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sciknoworg%2Fllm-atlas/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sciknoworg","download_url":"https://codeload.github.com/sciknoworg/llm-atlas/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sciknoworg%2Fllm-atlas/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28884285,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-29T19:55:09.949Z","status":"ssl_error","status_checked_at":"2026-01-29T19:55:08.490Z","response_time":59,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":"2026-01-29T20:17:01.234Z","updated_at":"2026-01-29T20:17:02.011Z","avatar_url":"https://github.com/sciknoworg.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Bachelor Thesis: Automated Knowledge Extraction for Large-Scale Generative AI Models Catalog\n\n[![CI](https://github.com/yourusername/Bachelor-Arbeit-NLP/workflows/CI/badge.svg)](https://github.com/yourusername/Bachelor-Arbeit-NLP/actions)\n[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\n## Overview\n\nThis project implements a Python-based NLP workflow that automatically extracts key information from Large Language Model (LLM) research papers and populates the ORKG comparison \"Generative AI Model Landscape\".\n\nThe pipeline supports both **API-based** extraction (using KISSKI Chat AI API) and **HPC cluster** deployment (using GPU-based transformers on GWDG Grete).\n\n## Features\n\n- **Automated paper fetching** from ArXiv\n- **PDF parsing** and text extraction with fallback mechanisms\n- **Dual extraction modes**:\n  - API-based using KISSKI Chat AI API (free for thesis work)\n  - GPU-based using Hugging Face transformers on HPC cluster\n- **Structured data mapping** to ORKG templates\n- **Automatic ORKG updates** with duplicate detection\n- **Multi-model extraction** from single papers\n- **Batch processing** support for HPC environments\n\n## Project Structure\n\n```\nBachelor-Arbeit-NLP/\n├── .github/\n│   └── workflows/\n│       └── ci.yml                    # GitHub Actions CI/CD\n├── docs/\n│   ├── deployment/\n│   │   ├── grete-setup.md           # HPC cluster setup guide\n│   │   └── verify-jobs.md           # Job verification guide\n│   ├── troubleshooting/\n│   │   ├── phi3-cache-issues.md     # Phi-3 troubleshooting\n│   │   └── distilgpt2-issues.md     # DistilGPT2 troubleshooting\n│   └── archive/\n│       └── cleanup-summary-2025-12.md\n├── grete/                            # HPC cluster deployment\n│   ├── extraction/\n│   │   ├── grete_extract_paper.py\n│   │   ├── grete_extract_from_url.py\n│   │   └── grete_extract_paper_distilgpt2.py\n│   ├── jobs/\n│   │   ├── grete_extract_job.sh     # SLURM single job\n│   │   ├── grete_extract_batch.sh   # SLURM batch array\n│   │   └── grete_extract_url_job.sh\n│   └── README.md                     # Grete-specific documentation\n├── src/                              # Core pipeline\n│   ├── __init__.py\n│   ├── comparison_updater.py         # Update ORKG comparisons\n│   ├── llm_extractor.py              # KISSKI API extraction\n│   ├── llm_extractor_transformers.py # GPU-based extraction\n│   ├── orkg_client.py                # ORKG API wrapper\n│   ├── orkg_manager.py               # ORKG management\n│   ├── paper_fetcher.py              # ArXiv paper fetching\n│   ├── pdf_parser.py                 # PDF text extraction\n│   ├── pipeline.py                   # Main orchestration\n│   └── template_mapper.py            # ORKG template mapping\n├── scripts/\n│   ├── append_to_paper.py            # Production utilities\n│   ├── sandbox_upload.py\n│   └── debug/                        # Debug/testing scripts\n│       ├── add_to_orkg_manual.py\n│       ├── export_to_csv.py\n│       └── force_new_paper_upload.py\n├── tests/                            # Unit tests\n│   ├── __init__.py\n│   ├── test_llm_extractor.py\n│   ├── test_orkg_append.py\n│   ├── test_orkg_client.py\n│   ├── test_pdf_parser.py\n│   └── test_pipeline.py\n├── examples/\n│   └── example_usage.py              # Usage examples\n├── data/\n│   ├── papers/                       # Downloaded PDFs\n│   ├── extracted/                    # Extracted JSON data\n│   └── logs/                         # Processing logs\n├── notebooks/\n│   └── validate_extraction.ipynb     # Validation notebook\n├── config/\n│   └── config.yaml                   # Configuration\n├── .env.example                      # Environment variables template\n├── .gitignore\n├── LICENSE                           # MIT License\n├── pyproject.toml                    # Modern Python packaging\n├── README.md\n├── requirements.txt                  # Production dependencies\n└── requirements-dev.txt              # Development dependencies\n```\n\n## Installation\n\n1. Clone the repository\n2. Create a virtual environment:\n   ```bash\n   python -m venv venv\n   source venv/bin/activate  # On Windows: venv\\Scripts\\activate\n   ```\n3. Install dependencies:\n   ```bash\n   pip install -r requirements.txt\n   ```\n4. Copy `.env.example` to `.env` and fill in your credentials:\n   ```bash\n   cp .env.example .env\n   ```\n\n## Configuration\n\n### Environment Variables (`.env`)\n\nRequired variables:\n```env\nORKG_EMAIL=your.email@example.com\nORKG_PASSWORD=your_password_here\nORKG_HOST=sandbox  # or 'production'\nKISSKI_API_KEY=your_kisski_api_key_here\nKISSKI_API_ENDPOINT=https://kisski.de/api/chat\n```\n\n### Pipeline Configuration (`config/config.yaml`)\n\nCustomize the extraction pipeline:\n- **ORKG settings**: host, template ID, comparison ID\n- **KISSKI API**: endpoint, model, temperature, max tokens, rate limiting\n- **ArXiv settings**: categories, download directory\n- **Extraction fields**: which fields to extract\n- **Logging**: level, format, output files\n\n## Usage\n\n### Basic Usage\n\n```python\nfrom src.pipeline import ExtractionPipeline\n\npipeline = ExtractionPipeline()\npipeline.process_paper(\"2302.13971\")  # ArXiv ID for Llama paper\n```\n\n### Command Line\n\n```bash\npython -m src.pipeline --arxiv-id 2302.13971\n```\n\n## Testing\n\n### Run All Tests\n\n```bash\npytest tests/ -v\n```\n\n### Run with Coverage\n\n```bash\npytest tests/ -v --cov=src --cov-report=html\n```\n\n### Run Specific Test File\n\n```bash\npytest tests/test_pipeline.py -v\n```\n\n### Continuous Integration\n\nThe project uses GitHub Actions for automated testing. See [`.github/workflows/ci.yml`](.github/workflows/ci.yml) for the CI configuration.\n\nTests run automatically on:\n- Push to `main` or `develop` branches\n- Pull requests to `main` or `develop`\n\n## Deployment\n\n### Local Extraction (API-based)\n\nUses KISSKI Chat AI API for extraction. Fast and reliable, free for thesis work.\n\n**Prerequisites**: KISSKI API key from your professor\n\nSee [Usage](#usage) section above for examples.\n\n### HPC Cluster (GPU-based)\n\nUses local transformers on GPU. No API costs, unlimited processing, but requires HPC setup.\n\n**Full deployment guide**: [docs/deployment/grete-setup.md](docs/deployment/grete-setup.md)\n\n**Quick start**:\n1. Upload code to Grete cluster\n2. Set up conda environment with PyTorch + transformers\n3. Submit SLURM jobs:\n   ```bash\n   sbatch grete/jobs/grete_extract_job.sh 2302.13971\n   ```\n\n**Monitoring**: [docs/deployment/verify-jobs.md](docs/deployment/verify-jobs.md)\n\n## Troubleshooting\n\nCommon issues and solutions:\n\n- **Phi-3 cache compatibility**: [docs/troubleshooting/phi3-cache-issues.md](docs/troubleshooting/phi3-cache-issues.md)\n- **DistilGPT2 JSON errors**: [docs/troubleshooting/distilgpt2-issues.md](docs/troubleshooting/distilgpt2-issues.md)\n- **Job verification**: [docs/deployment/verify-jobs.md](docs/deployment/verify-jobs.md)\n\n## Development\n\n### Code Quality\n\n```bash\n# Format code\nblack src/\n\n# Lint code\nflake8 src/\n\n# Type checking\nmypy src/\n\n# Sort imports\nisort src/\n```\n\n### Project Structure\n\n- **`src/`**: Core pipeline code\n- **`tests/`**: Unit tests\n- **`scripts/`**: Utility scripts\n- **`grete/`**: HPC cluster deployment files\n- **`docs/`**: Documentation\n- **`examples/`**: Usage examples\n- **`notebooks/`**: Jupyter notebooks for exploration\n\n### Contributing\n\n1. Create a feature branch\n2. Make your changes\n3. Run tests: `pytest tests/`\n4. Run linters: `black src/ \u0026\u0026 flake8 src/`\n5. Submit a pull request\n\n## Resources\n\n### ORKG\n\n- **Platform**: https://orkg.org/\n- **Sandbox**: https://sandbox.orkg.org/\n- **Python Client**: https://orkg.readthedocs.io/en/latest/\n- **LLM Template**: https://orkg.org/templates/R609825\n- **Target Comparison**: https://orkg.org/comparisons/R1364660\n\n### APIs\n\n- **KISSKI Chat AI API**: Provided by university for thesis work\n- **ArXiv API**: https://info.arxiv.org/help/api/\n\n### HPC\n\n- **GWDG Grete**: https://info.gwdg.de/docs/doku.php?id=en:services:application_services:high_performance_computing:grete\n\n## Citation\n\nIf you use this code for your research, please cite:\n\n```bibtex\n@thesis{bachelor2025llm,\n  title={Automated Knowledge Extraction for Large-Scale Generative AI Models Catalog},\n  author={Your Name},\n  year={2025},\n  school={Your University},\n  type={Bachelor's Thesis}\n}\n```\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## Acknowledgments\n\n- Developed as part of a Bachelor thesis project\n- ORKG platform for knowledge graph infrastructure\n- GWDG for providing HPC resources and KISSKI API access\n- SAIA platform for Chat AI API services\n\n## Contact\n\nFor questions or issues, please open an issue on GitHub or contact [your.email@example.com](mailto:your.email@example.com).\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsciknoworg%2Fllm-atlas","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsciknoworg%2Fllm-atlas","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsciknoworg%2Fllm-atlas/lists"}