trace_synthesis/tasks/559.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

40 lines
1.1 KiB
JSON

{
"sites": [
"gitlab",
"wikipedia"
],
"task_id": 559,
"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": "nolan_followers",
"topics": "career timeline of Christopher Nolan"
},
"intent": "Create a repo named nolan_followers with career timeline of 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/nolan_followers/-/raw/main/README.md",
"locator": "",
"required_contents": {
"must_include": [
"1993\u20132003: Early career and breakthrough",
"2003\u20132013: Widespread recognition",
"2014\u20132019: Established Hollywood auteur",
"2020\u2013present"
]
}
}
]
},
"intent_template_id": 87
}