trace_synthesis/tasks/586.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

46 lines
1.2 KiB
JSON

{
"sites": [
"shopping"
],
"task_id": 586,
"require_login": true,
"storage_state": "./.auth/shopping_state.json",
"start_url": "http://localhost:28082",
"geolocation": null,
"intent_template": "Rate my recent purchase of {{product}} with {{num_star}} stars, using my nickname {{nickname}}?",
"instantiation_dict": {
"product": "Jiffy Corn Muffin Cornbread Mix",
"num_star": 4,
"nickname": "ShoppingEmma"
},
"intent": "Rate my recent purchase of Jiffy Corn Muffin Cornbread Mix with 4 stars, using my nickname ShoppingEmma?",
"require_reset": false,
"eval": {
"eval_types": [
"program_html"
],
"reference_answers": null,
"reference_url": "",
"program_html": [
{
"url": "last",
"locator": "func:shopping_get_sku_latest_review_rating('B07HZB38XH')",
"required_contents": {
"must_include": [
"80"
]
}
},
{
"url": "last",
"locator": "func:shopping_get_sku_latest_review_author('B07HZB38XH')",
"required_contents": {
"must_include": [
"ShoppingEmma"
]
}
}
]
},
"intent_template_id": 194
}