trace_synthesis/tasks/236.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

37 lines
1014 B
JSON

{
"sites": [
"map"
],
"task_id": 236,
"require_login": true,
"storage_state": null,
"start_url": "https://www.openstreetmap.org",
"geolocation": null,
"intent_template": "Where is the nearest {{location}} from {{location2}} {{condition}}",
"instantiation_dict": {
"location": "pharmacy",
"location2": "Carnegie Mellon",
"condition": "I can walk within 20mins"
},
"intent": "Where is the nearest pharmacy from Carnegie Mellon I can walk within 20mins",
"require_reset": false,
"eval": {
"eval_types": [
"string_match"
],
"reference_answers": {
"must_include": [
"Schiller's Pharmacy",
"811",
"South Aiken Avenue",
"Shadyside",
"Pittsburgh"
]
},
"reference_url": "",
"program_html": [],
"string_note": "",
"reference_answer_raw_annotation": "Schiller's Pharmacy, 811, South Aiken Avenue, Shadyside, Pittsburgh, Allegheny County, 15232, United States"
},
"intent_template_id": 39
}