{"data":{"kind":"file","path":"README.md","version_id":"epgz3tqnuaoyj3kax18rax5k","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1275,"modified_at":"2026-02-08T18:43:18.862000","content_hash":"7096a7ef62c5372680f75e4e5e48104779fa7e0c974c2a971d08a2ce748571d6"},"entries":[],"content":"# OpenMed PMCPatients\n\nPatient case comprehension environment using the PMC-Patients dataset.\n\n## Task\n\nGiven a patient case description from PubMed Central medical literature, determine the patient's gender based on clinical text analysis. This tests reading comprehension and demographic extraction from unstructured medical text.\n\n## Dataset\n\n- **Source**: [zhengyun21/PMC-Patients](https://huggingface.co/datasets/zhengyun21/PMC-Patients)\n- **Size**: ~167,000 patient case summaries\n- **Format**: Patient case description -> gender classification (male/female)\n- **License**: Open\n\n## Reward Structure\n\n| Reward Function | Weight | Description |\n|----------------|--------|-------------|\n| accuracy_reward | 45% | Exact gender match |\n| partial_match_reward | 20% | Credit for valid gender prediction |\n| thinking_reward | 20% | Quality of clinical text analysis |\n| format_reward | 15% | Proper \\\\boxed{} or \\|answer\\| format |\n\n## Example\n\n**Input**: \"A 45-year-old patient presented with chest pain. He reported the pain started 2 hours ago...\"\n\n**Expected Output**: `\\boxed{male}`\n\n## Citation\n\n```bibtex\n@article{zhao2023pmcpatients,\n  title={Large-scale Dataset of Real-world Patients from PubMed Central},\n  author={Zhao, Zhengyun and others},\n  year={2023}\n}\n```\n","encoding":"utf-8","truncated":false,"total_bytes":1275},"status":null}