Back to Prompt Optimizer

pro-multi-compare

docs/workspace/compare-evaluation-analysis/real-api-samples/pro-multi-compare/response.md

2.10.22.2 KB
Original Source

pro-multi-compare

Meta

  • Type: compare
  • Model: dashscope
  • Duration: 10268ms
  • Timestamp: 1773729237468

Score

  • Overall: 60
KeyLabelScore
goalAchievementRobustness目标达成稳定性40
outputQualityCeiling输出质量上限70
promptPatternQuality提示词模式质量50
crossSnapshotRobustness跨快照鲁棒性30
workspaceTransferability对工作区的可迁移性80

Improvements

  • 明确要求 assistant 在提供建议前必须先澄清用户需求,避免直接给出方案。
  • 统一使用更结构化的引导语句(如‘请先确认以下几点’)来确保行为一致性。
  • 在提示词中加入具体问题示例,以帮助模型更好地理解如何逐步推进对话。

Summary

当前 system 消息未能稳定促使 assistant 先澄清用户目标,需加强约束与引导。

Raw Response JSON

json
{
  "type": "compare",
  "score": {
    "overall": 60,
    "dimensions": [
      {
        "key": "goalAchievementRobustness",
        "label": "目标达成稳定性",
        "score": 40
      },
      {
        "key": "outputQualityCeiling",
        "label": "输出质量上限",
        "score": 70
      },
      {
        "key": "promptPatternQuality",
        "label": "提示词模式质量",
        "score": 50
      },
      {
        "key": "crossSnapshotRobustness",
        "label": "跨快照鲁棒性",
        "score": 30
      },
      {
        "key": "workspaceTransferability",
        "label": "对工作区的可迁移性",
        "score": 80
      }
    ]
  },
  "improvements": [
    "明确要求 assistant 在提供建议前必须先澄清用户需求,避免直接给出方案。",
    "统一使用更结构化的引导语句(如‘请先确认以下几点’)来确保行为一致性。",
    "在提示词中加入具体问题示例,以帮助模型更好地理解如何逐步推进对话。"
  ],
  "summary": "当前 system 消息未能稳定促使 assistant 先澄清用户目标,需加强约束与引导。",
  "patchPlan": [],
  "metadata": {
    "model": "dashscope",
    "timestamp": 1773729237468,
    "duration": 10268
  }
}