trace_synthesis/tasks/561.json
yuyr a84d51a101 1. 增加r1生成综合策略代码和输出;
2. 增加tasks;
3. 增加analysis部分,对策略进行归纳分类,然后进行评测。
2025-04-17 17:40:15 +08:00

43 lines
1.1 KiB
JSON

{
"sites": [
"gitlab",
"wikipedia"
],
"task_id": 561,
"require_login": true,
"storage_state": "./.auth/gitlab_state.json",
"start_url": "http://localhost:28084",
"geolocation": null,
"intent_template": "Create a repo named {{name}} with {{topics}} in a README file",
"instantiation_dict": {
"name": "bafta_awards_nolan",
"topics": "movies that are nominated BAFTA Awards by Christopher Nolan"
},
"intent": "Create a repo named bafta_awards_nolan with movies that are nominated BAFTA Awards by Christopher Nolan in a README file",
"require_reset": false,
"eval": {
"eval_types": [
"program_html"
],
"reference_answers": null,
"reference_url": "",
"program_html": [
{
"url": "http://localhost:28084/byteblaze/bafta_awards_nolan/-/raw/main/README.md",
"locator": "",
"required_contents": {
"must_include": [
"Batman Begins",
"The Dark Knight",
"Inception",
"The Dark Knight Rises",
"Interstellar",
"Dunkirk",
"Tenet"
]
}
}
]
},
"intent_template_id": 87
}