兩項目評估報告
- EvalRun: #60(suite #36
auto-4f699cec-r1-053958385)
- Target: #2 production-canary
- Cases: 30(N=1, attempts=30/30)
- Generated: 2026-05-16T12:25:41+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 |
❌ 25.3% |
✅ 100.0% |
待修方向(worst-3)
- 知識與產品查詢 × retrieval_relevance = 25.3% — 撈到的 KB chunks 不夠相關 — KB content gap,跟 Neptune team 提 enrichment
- 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 |
6 |
30.0% |
| Step 2: + tool_calling |
6 |
30.0% |
| Step 3: + answer |
2 |
10.0% |
Per-scenario funnel
| Scenario (n_attempts) |
Step 1 (routing) |
Step 2 (tool) |
Step 3 (answer) |
end-to-end |
| ALTRA - 商品推薦 (2) |
⚪ 1/2 (50.0%) |
⚪ 1/2 (50.0%) |
❌ 0/2 (0.0%) |
❌ 0/2 (0.0%) |
| early_return (9) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
❌ 0/9 (0.0%) |
| 庫存查詢 (3) |
❌ 1/3 (33.3%) |
❌ 1/3 (33.3%) |
❌ 0/3 (0.0%) |
❌ 0/3 (0.0%) |
| 物流狀態查詢 (3) |
⚪ 2/3 (66.7%) |
⚪ 2/3 (66.7%) |
❌ 1/3 (33.3%) |
❌ 1/3 (33.3%) |
| 知識與產品查詢 (2) |
✅ 2/2 (100.0%) |
✅ 2/2 (100.0%) |
⚪ 1/2 (50.0%) |
⚪ 1/2 (50.0%) |
| 退換貨申請 (1) |
❌ 0/1 (0.0%) |
❌ 0/1 (0.0%) |
❌ 0/1 (0.0%) |
❌ 0/1 (0.0%) |
Drop-off 最大的 5 個 scenario
- early_return drop at step1 (-100.0pp)
- 退換貨申請 drop at step1 (-100.0pp)
- 庫存查詢 drop at step1 (-66.7pp)
- ALTRA - 商品推薦 drop at step1 (-50.0pp)
- 知識與產品查詢 drop at step3 (-50.0pp)
Audit
Reproduce combined report: bin/rails runner "puts Eval::EvaluationReport.call(run: EvalRun.find(60))"
Or fetch each item separately:
bin/rails runner "puts Eval::KbAccuracyReport.call(run: EvalRun.find(60))"
bin/rails runner "puts Eval::ScenarioFunnelReport.call(run: EvalRun.find(60))"
Per-Scenario × Per-Dim — Run #60
Suite: ALTRA客服小幫手 (bulk R1) · scenarios: 6 · dims: 5 · populated cells: 29/30
| Scenario |
Scenario |
Tool |
Retrieval |
Faith |
AnsQ |
| ALTRA - 商品推薦 |
100.0% [100.0–100.0] (n=2) |
100.0% [100.0–100.0] (n=2) |
50.0% [0.0–100.0] (n=2) |
100.0% [100.0–100.0] (n=2) |
76.7% [73.3–80.0] (n=2) |
| early_return |
0.0% [0.0–0.0] (n=9) |
0.0% (n=1) |
87.5% [62.5–100.0] (n=8) |
94.4% [83.3–100.0] (n=9) |
90.7% [76.7–99.3] (n=9) |
| 庫存查詢 |
100.0% [100.0–100.0] (n=3) |
100.0% [100.0–100.0] (n=3) |
0.0% [0.0–0.0] (n=3) |
0.0% [0.0–0.0] (n=3) |
30.0% [0.0–90.0] (n=3) |
| 物流狀態查詢 |
100.0% [100.0–100.0] (n=3) |
0.0% [0.0–0.0] (n=3) |
— |
0.0% [0.0–0.0] (n=3) |
83.3% [50.0–100.0] (n=3) |
| 知識與產品查詢 |
100.0% [100.0–100.0] (n=12) |
100.0% [100.0–100.0] (n=12) |
90.0% [70.0–100.0] (n=10) |
88.9% [69.4–100.0] (n=12) |
91.7% [79.4–100.0] (n=12) |
| 退換貨申請 |
100.0% (n=1) |
0.0% (n=1) |
0.0% (n=1) |
0.0% (n=1) |
90.0% (n=1) |
Worst-3 cells (lowest primary score)
- 物流狀態查詢 × Tool · 0.0% (n=3) · lowest sub_metric:
tools_recall
- 物流狀態查詢 × Faith · 0.0% (n=3) · lowest sub_metric:
rule_compliance
- 庫存查詢 × Faith · 0.0% (n=3) · lowest sub_metric:
rule_compliance