{"data":{"kind":"file","path":"README.md","version_id":"fgans64ympu5gugrcryr1o0d","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1057,"modified_at":"2025-12-03T04:59:56.155000","content_hash":"f1fa0d730805496629580806673fc0881ef5171cb8820896f2eb9feffeca387a"},"entries":[],"content":"# sentence-repeater\n\n### Overview\n- **Environment ID**: `sentence-repeater`\n- **Short description**: Multi-turn QA over a paragraph to retrieve specific sentences in random order.\n- **Tags**: retrieval, multi-turn, text, similarity\n\n### Datasets\n- **Primary dataset(s)**: `agentlans/wikipedia-paragraphs` processed into 5 Q/A turns per paragraph\n- **Source links**: Hugging Face Datasets\n- **Split sizes**: Train split filtered to paragraphs with 5 adequate-length sentences\n\n### Task\n- **Type**: multi-turn\n- **Parser**: default `Parser`\n- **Rubric overview**: Sequence similarity against the 5 target answers across turns\n\n### Quickstart\nRun an evaluation with default settings:\n\n```bash\nuv run vf-eval sentence-repeater\n```\n\nConfigure model and sampling:\n\n```bash\nuv run vf-eval sentence-repeater \\\n  -m gpt-4.1-mini \\\n  -n 20 -r 3 -t 1024 -T 0.7\n```\n\n\n### Environment Arguments\nThis loader does not expose custom arguments.\n\n### Metrics\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Mean SequenceMatcher ratio across the 5 repeated sentences |\n","encoding":"utf-8","truncated":false,"total_bytes":1057},"status":null}