docs: wiki vs RAG distinction in model lifecycle
wiki is not traditional RAG: - RAG: ephemeral query-time augmentation - Wiki: permanent model corrections, versioned, packaged with model - Edits accumulate across versions as ground truth
This commit is contained in:
@@ -119,42 +119,75 @@ release/
|
||||
|
||||
每個 momentry model(`{uuid}_v1` / `v2` / `v3`)不只是唯讀的產出,而是可透過 wiki 機制持續改善。
|
||||
|
||||
### 與傳統 RAG 的區別
|
||||
|
||||
| | 傳統 RAG | momentry wiki |
|
||||
|---|---|---|
|
||||
| 知識儲存 | vector DB(ephemeral) | model package(permanent) |
|
||||
| 修正方式 | query 時 LLM 決定是否採用 | 使用者/Agent 直接編輯 |
|
||||
| 修正持久性 | ❌ 下次 query 就消失 | ✅ 寫入 model,版本化保存 |
|
||||
| 模型改進 | 無(僅改變 prompt) | 下次 version bump 時合併為 ground truth |
|
||||
| 協作方式 | 單向(retrieve → generate) | 雙向(編輯 → 合併 → 改進) |
|
||||
| 離線可用 | ❌ 需 vector DB + LLM | ✅ 離線查閱 wiki 目錄 |
|
||||
|
||||
**momentry wiki 不是 RAG 的替代品,而是 model 的生命週期管理機制。**
|
||||
|
||||
### 概念
|
||||
|
||||
```
|
||||
momentry model (release package)
|
||||
├── output_json/ ← 唯讀,processor 產出
|
||||
├── chunks/ ← 唯讀,ingestion 產出
|
||||
├── vectors/ ← 唯讀,embedding 產出
|
||||
├── chunks.csv ← 唯讀,ingestion 產出
|
||||
├── vectors.csv ← 唯讀,embedding 產出
|
||||
└── wiki/ ← 可編輯,使用者貢獻知識
|
||||
├── identities.json ← "這個 trace 是 Audrey Hepburn"
|
||||
├── objects.json ← "這個 object 是郵票 #42"
|
||||
├── corrections.json ← "ASR 'Hello' 應為 'Halo'"
|
||||
├── identities.json ← "trace 5 = Audrey Hepburn"
|
||||
├── objects.json ← "object 42 = 郵票 #1"
|
||||
├── corrections.json ← "ASR 'Hello' → 'Halo'"
|
||||
└── changelog.json ← 編輯歷史
|
||||
```
|
||||
|
||||
### wiki 資料流向
|
||||
### 資料流向
|
||||
|
||||
```
|
||||
使用者/Agent 編輯 wiki
|
||||
↓
|
||||
wiki/{topic}.json 更新
|
||||
DB wiki_entries + wiki_revisions 寫入
|
||||
↓
|
||||
下次 release 打包時合併
|
||||
下次 release 打包時 merge 進 model
|
||||
↓
|
||||
新版 model(version bump)
|
||||
TKG label 更新 (tkg_nodes.label)
|
||||
↓
|
||||
新版 model version bump
|
||||
```
|
||||
|
||||
### 與 TKG 的關係
|
||||
|
||||
wiki 的 identity 和 object 標註會回寫到 TKG:
|
||||
wiki 的 identity 和 object 標註會回寫到 TKG node label:
|
||||
```
|
||||
(face_trace:5) -[:IDENTIFIED_AS]-> (identity:Audrey Hepburn) ← wiki 編輯
|
||||
(object_instance:42) -[:LABELED_AS]-> (object_class:stamp) ← wiki 編輯
|
||||
(face_trace:5) label="Audrey Hepburn" ← wiki 編輯
|
||||
(object_instance:42) label="郵票 #1" ← wiki 編輯
|
||||
```
|
||||
|
||||
這些編輯累積後,可做為下一版 model training 的 ground truth。
|
||||
|
||||
### 實作方向
|
||||
|
||||
**DB 層** — 新 table `wiki_entries` + `wiki_revisions`:
|
||||
```sql
|
||||
wiki_entries (target_type, target_id, title, body, summary, status, version, file_uuid)
|
||||
wiki_revisions (entry_id, version, title, body, summary, change_summary, edited_by)
|
||||
```
|
||||
|
||||
**API 層** — CRUD + 版本歷史:
|
||||
```
|
||||
GET /api/v1/wiki/{target_type}/{target_id}
|
||||
PUT /api/v1/wiki/{target_type}/{target_id}
|
||||
GET /api/v1/wiki/{target_type}/{target_id}/revisions
|
||||
POST /api/v1/wiki/search
|
||||
```
|
||||
|
||||
**打包層** — `release_pack.py` 加入 wiki 匯出,與 model 共存
|
||||
|
||||
---
|
||||
|
||||
## Phase 3:Object Identity(v3 model)
|
||||
|
||||
Reference in New Issue
Block a user