trace_synthesis/tasks/325.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

27 lines
791 B
JSON

{
"sites": [
"shopping"
],
"task_id": 325,
"require_login": true,
"storage_state": "./.auth/shopping_state.json",
"start_url": "http://localhost:28082",
"geolocation": null,
"intent_template": "Show me the \"{{product}}\" listings by {{sorting_order}}.",
"instantiation_dict": {
"product": "mouth night guard",
"sorting_order": "descending price"
},
"intent": "Show me the \"mouth night guard\" listings by descending price.",
"require_reset": false,
"eval": {
"eval_types": [
"url_match"
],
"reference_answers": null,
"reference_url": "http://localhost:28082/catalogsearch/result/index/?q=mouth%20night%20guard%20&product_list_order=price",
"program_html": [],
"url_note": "GOLD in PRED"
},
"intent_template_id": 208
}