Record the phase05 live eval and result summary
This commit is contained in:
File diff suppressed because one or more lines are too long
@@ -0,0 +1,320 @@
|
||||
{
|
||||
"target": "http://127.0.0.1:18817",
|
||||
"model": "MiniMax-M2.7-highspeed",
|
||||
"generatedAt": "2026-10-10T00:08:03Z",
|
||||
"tasks": [
|
||||
{
|
||||
"id": "minutes-basic",
|
||||
"turns": [
|
||||
{
|
||||
"user": "帮我总结一份会议纪要",
|
||||
"runId": "run_98227b6da4f8",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_98227b6da4f8",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:08:03.365884236Z",
|
||||
"endedAt": "2026-10-10T00:09:46.817340834Z",
|
||||
"usage": {
|
||||
"promptTokens": 958971,
|
||||
"completionTokens": 5805,
|
||||
"totalTokens": 964776,
|
||||
"durationMs": 103450,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 20,
|
||||
"toolFailures": 2,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 1,
|
||||
"planUpdates": 6,
|
||||
"verifierChecks": 2
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "事实核查:交付物里以下数字/日期在本次材料中找不到出处:2026年、10月10日。逐条二选一:改成材料里能找到的数值,或删除/明确标注为推测或目标值(如“目标 85%”)。不要保留无法回溯的裸数字。"
|
||||
},
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"artifacts": [
|
||||
"研发智能体文档处理与自举方案评审纪要.docx",
|
||||
"研发智能体文档处理与自举方案评审纪要.docx"
|
||||
],
|
||||
"durationMs": 105127
|
||||
}
|
||||
],
|
||||
"total": {
|
||||
"toolCalls": 20,
|
||||
"toolFailures": 2,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 1,
|
||||
"planUpdates": 6,
|
||||
"verifierChecks": 2
|
||||
},
|
||||
"unverifiedFacts": [
|
||||
"10月10日"
|
||||
]
|
||||
},
|
||||
{
|
||||
"id": "plan-followup",
|
||||
"turns": [
|
||||
{
|
||||
"user": "帮我总结一份会议纪要",
|
||||
"runId": "run_0e2939910b9b",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_0e2939910b9b",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:09:48.529064498Z",
|
||||
"endedAt": "2026-10-10T00:11:19.797414578Z",
|
||||
"usage": {
|
||||
"promptTokens": 490283,
|
||||
"completionTokens": 5397,
|
||||
"totalTokens": 495680,
|
||||
"durationMs": 91268,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 16,
|
||||
"toolFailures": 4,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 1,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 5,
|
||||
"verifierChecks": 1
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"artifacts": [
|
||||
"研发智能体文档处理与自举方案评审会议纪要.docx"
|
||||
],
|
||||
"durationMs": 93112
|
||||
},
|
||||
{
|
||||
"user": "根据这个纪要,制定一个方案",
|
||||
"runId": "run_db60273d0c50",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_db60273d0c50",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:11:21.632969359Z",
|
||||
"endedAt": "2026-10-10T00:12:41.942635887Z",
|
||||
"usage": {
|
||||
"promptTokens": 527953,
|
||||
"completionTokens": 4448,
|
||||
"totalTokens": 532401,
|
||||
"durationMs": 80309,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 12,
|
||||
"toolFailures": 1,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 2,
|
||||
"verifierChecks": 2
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "事实核查:交付物里以下数字/日期在本次材料中找不到出处:2026-10-10。逐条二选一:改成材料里能找到的数值,或删除/明确标注为推测或目标值(如“目标 85%”)。不要保留无法回溯的裸数字。"
|
||||
},
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"artifacts": [
|
||||
"研发智能体文档处理系统重构方案.docx",
|
||||
"研发智能体文档处理系统重构方案.docx"
|
||||
],
|
||||
"durationMs": 81106
|
||||
},
|
||||
{
|
||||
"user": "方案太简要了,给我一个详细的方案和计划",
|
||||
"runId": "run_aa3daf21b2bc",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_aa3daf21b2bc",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:12:42.741977532Z",
|
||||
"endedAt": "2026-10-10T00:16:13.009298846Z",
|
||||
"usage": {
|
||||
"promptTokens": 2124640,
|
||||
"completionTokens": 15096,
|
||||
"totalTokens": 2139736,
|
||||
"durationMs": 210267,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 30,
|
||||
"toolFailures": 5,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 2,
|
||||
"verifierChecks": 2
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "事实核查:交付物里以下数字/日期在本次材料中找不到出处:95%、80%、90%、85%、20%、70%。逐条二选一:改成材料里能找到的数值,或删除/明确标注为推测或目标值(如“目标 85%”)。不要保留无法回溯的裸数字。"
|
||||
},
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"artifacts": [
|
||||
"研发智能体文档处理系统重构详细方案.docx",
|
||||
"研发智能体文档处理系统重构详细方案.docx"
|
||||
],
|
||||
"durationMs": 210280
|
||||
}
|
||||
],
|
||||
"total": {
|
||||
"toolCalls": 58,
|
||||
"toolFailures": 10,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 1,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 9,
|
||||
"verifierChecks": 5
|
||||
},
|
||||
"unverifiedFacts": [
|
||||
"95%",
|
||||
"80%",
|
||||
"90%",
|
||||
"85%",
|
||||
"20%",
|
||||
"70%"
|
||||
]
|
||||
},
|
||||
{
|
||||
"id": "qa-citation",
|
||||
"turns": [
|
||||
{
|
||||
"user": "这次会议最主要的事情是什么",
|
||||
"runId": "run_c23966341ee0",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_c23966341ee0",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:16:13.059935626Z",
|
||||
"endedAt": "2026-10-10T00:16:29.456464692Z",
|
||||
"usage": {
|
||||
"promptTokens": 23148,
|
||||
"completionTokens": 728,
|
||||
"totalTokens": 23876,
|
||||
"durationMs": 16396,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 0,
|
||||
"toolFailures": 0,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 0,
|
||||
"verifierChecks": 1
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"durationMs": 18029
|
||||
}
|
||||
],
|
||||
"total": {
|
||||
"toolCalls": 0,
|
||||
"toolFailures": 0,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 0,
|
||||
"verifierChecks": 1
|
||||
}
|
||||
},
|
||||
{
|
||||
"id": "minutes-pm-tool",
|
||||
"turns": [
|
||||
{
|
||||
"user": "生成会议纪要",
|
||||
"runId": "run_e87920a123a3",
|
||||
"status": "completed",
|
||||
"finalText": "本轮未做同源模型验证(验证器未独立配置);硬事实与事实核查已执行",
|
||||
"usage": {
|
||||
"id": "run_e87920a123a3",
|
||||
"status": "completed",
|
||||
"createdAt": "2026-10-10T00:16:31.112896348Z",
|
||||
"endedAt": "2026-10-10T00:18:24.614591662Z",
|
||||
"usage": {
|
||||
"promptTokens": 647994,
|
||||
"completionTokens": 7187,
|
||||
"totalTokens": 655181,
|
||||
"durationMs": 113501,
|
||||
"toolCalls": 0
|
||||
},
|
||||
"error": null
|
||||
},
|
||||
"metrics": {
|
||||
"toolCalls": 19,
|
||||
"toolFailures": 7,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 0,
|
||||
"verifierChecks": 1
|
||||
},
|
||||
"checks": [
|
||||
{
|
||||
"passed": false,
|
||||
"reason": "验证器未配置独立轻量模型(当前与主模型相同),跳过同源验证"
|
||||
}
|
||||
],
|
||||
"artifacts": [
|
||||
"会议纪要-智能体项目管理工具演示与结题报告推进.docx"
|
||||
],
|
||||
"durationMs": 114121
|
||||
}
|
||||
],
|
||||
"total": {
|
||||
"toolCalls": 19,
|
||||
"toolFailures": 7,
|
||||
"fullRegens": 0,
|
||||
"apiDiscovery": 0,
|
||||
"readsOriginal": 0,
|
||||
"planUpdates": 0,
|
||||
"verifierChecks": 1
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
Reference in New Issue
Block a user