{"data":{"kind":"file","path":"README.md","version_id":"g8ozxicoyol2923vs5f8guea","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1026,"modified_at":"2025-08-22T01:50:44","content_hash":"02b6aebdc47e8808de5f4aac41e76d192fc35c6bcd4ec341a01f393c1a7f3272"},"entries":[],"content":"# xlam-function-calling\n\n### Overview\n- **Environment ID**: `xlam-function-calling`\n- **Short description**: Function-calling reproduction from XLAM-60k: model must emit a JSON array of tool calls.\n\n### Datasets\n- **Primary dataset(s)**: `Salesforce/xlam-function-calling-60k` (train split)\n- **Source links**: Hugging Face Datasets\n- **Split sizes**: Uses the `train` split for training/evaluation\n\n### Task\n- **Type**: single-turn\n- **Parser**: `XMLParser([\"think\",\"tool\"], answer_field=\"tool\")`\n- **Rubric overview**: Exact set equality between parsed JSON array and target tool list\n\n### Quickstart\nRun an evaluation with default settings:\n\n```bash\nuv run vf-eval xlam-function-calling\n```\n\nConfigure model and sampling:\n\n```bash\nuv run vf-eval xlam-function-calling \\\n  -m gpt-4.1-mini \\\n  -n 20 -r 3 -t 1024 -T 0.7\n```\n\n### Environment Arguments\nThis loader does not expose custom arguments.\n\n### Metrics\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | 1.0 if parsed tool JSON exactly matches target, else 0.0 |\n\n","encoding":"utf-8","truncated":false,"total_bytes":1026},"status":null}