兩項目評估報告
- EvalRun: #238(suite #131
auto-bb68782a-r1-053959635)
- Target: #2 production-canary
- Cases: 30(N=1, attempts=30/30)
- Generated: 2026-05-16T13:52:09+08:00
本報告含兩個獨立評估項目。項目一 評估 bot 進入「知識與產品查詢」fallback
後的撈取 + 回答能力;項目二 評估 bot 跨全 enabled scenarios 的 routing →
tool → answer 三段 funnel 健康度。
項目一: 知識庫精準度
評估範圍 (prerequisite: 進對 scenario)
- 進對 scenario 的 attempts: 10 / 30 (33.3%)
- 只對這些 attempts 算 retrieval + answer
- 沒進對 scenario 的 attempts → 排除(routing 失敗不該污染 KB 訊號)
Per-scenario 評估
| Scenario |
qualifying / total attempts |
retrieval_relevance |
answer_correctness |
| 知識與產品查詢 |
10 / 10 |
⚪ 54.3% |
✅ 100.0% |
待修方向(worst-3)
- 知識與產品查詢 × retrieval_relevance = 54.3% — retrieval 略低 — 看 LLM relevance vs topic_match 哪邊弱
- 知識與產品查詢 × answer_correctness = 100.0% — 回答內容有問題 — 細看 hallucination_rate vs answer_quality 哪邊低
項目二: 情境調用與完成
整體 funnel(全 scenarios 加總)
| Stage |
Pass count |
% of total |
| Total attempts |
20 |
100.0% |
| Step 1: scenario_routing |
5 |
25.0% |
| Step 2: + tool_calling |
5 |
25.0% |
| Step 3: + answer |
2 |
10.0% |
Per-scenario funnel
| Scenario (n_attempts) |
Step 1 (routing) |
Step 2 (tool) |
Step 3 (answer) |
end-to-end |
| early_return (9) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
| 月租諮詢 (5) |
❌ 1/5 (20.0%) |
❌ 1/5 (20.0%) |
❌ 1/5 (20.0%) |
❌ 1/5 (20.0%) |
| 特約合作 (2) |
⚪ 1/2 (50.0%) |
⚪ 1/2 (50.0%) |
⚪ 1/2 (50.0%) |
⚪ 1/2 (50.0%) |
| 知識與產品查詢 (4) |
⚪ 3/4 (75.0%) |
⚪ 3/4 (75.0%) |
❌ 0/4 (0.0%) |
❌ 0/4 (0.0%) |
Drop-off 最大的 5 個 scenario
- early_return drop at step1 (-100.0pp)
- 月租諮詢 drop at step1 (-80.0pp)
- 知識與產品查詢 drop at step3 (-75.0pp)
- 特約合作 drop at step1 (-50.0pp)
Audit
Reproduce combined report: bin/rails runner "puts Eval::EvaluationReport.call(run: EvalRun.find(238))"
Or fetch each item separately:
bin/rails runner "puts Eval::KbAccuracyReport.call(run: EvalRun.find(238))"
bin/rails runner "puts Eval::ScenarioFunnelReport.call(run: EvalRun.find(238))"
Per-Scenario × Per-Dim — Run #238
Suite: Times Parking AI 小幫手 (bulk R1) · scenarios: 4 · dims: 5 · populated cells: 20/20
| Scenario |
Scenario |
Tool |
Retrieval |
Faith |
AnsQ |
| early_return |
0.0% [0.0–0.0] (n=9) |
0.0% (n=1) |
62.5% [25.0–87.5] (n=8) |
100.0% [100.0–100.0] (n=6) |
65.2% [41.1–87.4] (n=9) |
| 月租諮詢 |
40.0% [0.0–80.0] (n=5) |
0.0% [0.0–0.0] (n=5) |
25.0% [0.0–50.0] (n=4) |
80.0% [40.0–100.0] (n=5) |
62.7% [30.7–81.3] (n=5) |
| 特約合作 |
50.0% [0.0–100.0] (n=2) |
50.0% [0.0–100.0] (n=2) |
50.0% [0.0–100.0] (n=2) |
16.7% [0.0–33.3] (n=2) |
55.0% [26.7–83.3] (n=2) |
| 知識與產品查詢 |
92.9% [78.6–100.0] (n=14) |
92.9% [78.6–100.0] (n=14) |
100.0% [100.0–100.0] (n=10) |
72.2% [44.4–91.7] (n=12) |
79.3% [65.0–92.1] (n=14) |
Worst-3 cells (lowest primary score)
- early_return × Scenario · 0.0% (n=9) · lowest sub_metric:
scenario_precision
- early_return × Tool · 0.0% (n=1) · lowest sub_metric:
tools_precision
- 月租諮詢 × Tool · 0.0% (n=5) · lowest sub_metric:
tools_recall