智能客服 - 犀犀

production env · 1 suites · 2 次完成 run

Subject № 5c7c4a1f-bde1-4109-8f36-437a6d7f3557 PRODUCTION

評測狀態 · 待重生

6 scenarios · 376 KB items 1 suite (1 stale)

FLEET READY · ALL KINDS COVERED

kb_accuracy 10

scenario_funnel 10

mixed_qa 10

uncategorized 0

生命徵象

[KIND × DIMENSION] vital signs — this bot's per-dim clearance vs. its baseline

知識庫精準度 [FAIL]

檢索 60.0% 60.0% <70.0% [floor] -10.0 ✗

忠實度 100.0% 85.7% <92.0% [±8pp] -6.3 ✗

回答品質 65.3% 64.3% <70.0% [floor] -5.7 ✗

情境調用與完成 [PASS]

情境 100.0% 100.0% ≥95.0% [±5pp] +5.0 ✓

工具使用 0.0% 0.0% — —

回答品質 84.7% 83.7% ≥76.7% [±8pp] +7.0 ✓

對話素養（混合問答） [FAIL]

檢索 100.0% 88.9% <95.0% [±5pp] -6.1 ✗

忠實度 96.7% 100.0% ≥88.7% [±8pp] +11.3 ✓

回答品質 96.0% 94.7% ≥88.0% [±8pp] +6.7 ✓