trace_synthesis/tasks/760.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

54 lines
1.4 KiB
JSON

{
"sites": [
"map",
"shopping_admin"
],
"task_id": 760,
"require_login": true,
"storage_state": null,
"start_url": "https://www.openstreetmap.org",
"geolocation": null,
"intent_template": "Show me the route and driving time from {{city1}} to {{city2}}",
"instantiation_dict": {
"city1": "Allentown, PA",
"city2": "the city where my E-commerce customer Amanda Kim lives"
},
"intent": "Show me the route and driving time from Allentown, PA to the city where my E-commerce customer Amanda Kim lives",
"require_reset": false,
"eval": {
"eval_types": [
"program_html"
],
"reference_answers": null,
"reference_url": "",
"program_html": [
{
"url": "last",
"locator": "document.querySelector(\"div#content select.routing_engines\").selectedIndex",
"required_contents": {
"exact_match": "1"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"route_from\"').value",
"required_contents": {
"must_include": [
"Allentown"
]
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"route_to\"').value",
"required_contents": {
"must_include": [
"Hoboken",
"New Jersey"
]
}
}
]
},
"intent_template_id": 42
}