36 Commits
Author SHA1 Message Date
imacatandClaude Fable 5 d46c50a2db Changed the natural-coding definition files to single-pass convergence
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 964006ec6d Add the natural-coding definition files and the convergence algorithms
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 0c82da7a1d Settle women-power as the a-priori screen label term
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 93a4f808bf Reduce run-llm to a pure batch executor
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 7a58f37d9f Record the natural-coding pipeline design in the project documents
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 b727284ef3 Gather the CLI command modules into a commands sub-package
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 396848e898 Add the export-llm-input subcommand for the lyrics-only LLM input
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 55926b8dc4 Regenerate the artists report with the Wikidata snapshot applied
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:26 +08:00
imacatandClaude Fable 5 ee6a1b88c5 Rename the Markdown documents to dash-connected names
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-17 22:38:20 +08:00
imacatandClaude Fable 5 1a7519becd Drop the engine preparation from build-db
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:36 +08:00
imacatandClaude Fable 5 a534b1b70b Move the invariant checks into the importers and drop find_violations
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:36 +08:00
imacatandClaude Fable 5 8fdd21f8a2 Package the capture-layer application into a CaptureImporter class
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:36 +08:00
imacatandClaude Fable 5 c561330bd0 Repackage the build-db import steps into job classes
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:36 +08:00
imacatandClaude Fable 5 64c62eb394 Capture the song lyrics from the pilot study and the public APIs
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:36 +08:00
imacatandClaude Fable 5 0d3cd69a0f Align the fetch-lyrics summary with fetch-artists via shared utils
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:34:26 +08:00
imacatandClaude Fable 5 81975b72e4 Remove the missing-lyrics report from fetch-lyrics
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:15:49 +08:00
imacatandClaude Fable 5 1b21471845 Fall back to the original artist credit when fetching lyrics
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:15:49 +08:00
imacatandClaude Fable 5 15f207ffc6 Remove the retired artist overrides layer and the mixed artist type
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:15:49 +08:00
imacat 93aa06703b Keep the artist snapshot sorted by name 2026-08-04 15:15:49 +08:00
imacat 8fdd25f569 Show resolved count and elapsed time in the fetch-artists summary 2026-08-04 15:15:49 +08:00
imacat 00b5c208ec Resolve artists on Wikidata via exact-match SPARQL with song corroboration 2026-08-04 15:15:49 +08:00
imacat c81bfcf51d Restore official artist name spellings in the canonical table 2026-08-04 15:15:49 +08:00
imacat 942a74c86c Record the complete-Wikidata-then-capture workflow decision 2026-08-04 15:15:49 +08:00
imacat 1f95f0231d Export human-readable songs and artists reports to the derived data layer 2026-08-04 15:15:49 +08:00
imacat ae0e9d0a08 Split artist credits with documented exceptions and normalize artist names 2026-08-04 15:15:49 +08:00
imacat 6a91bc73eb Merge same-song chart rows via a canonical artist credit table 2026-08-04 15:15:49 +08:00
imacat e15c988bf4 Store the source chart CSV with its original CRLF line endings 2026-08-04 15:15:49 +08:00
imacat f36a30411d Align the source chart CSV with the archived Billboard year-end charts 2026-08-04 15:15:49 +08:00
imacatandClaude Fable 5 5d5e4fac50 Reorganize data/ into source, captures, and manual layers
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:15:49 +08:00
imacatandClaude Fable 5 e9027472f4 Make every subcommand input an explicit CLI argument or option
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:13:31 +08:00
imacatandClaude Fable 5 e0eb343ce0 Refactor internal data passing to dataclasses and enums, and remove unneeded defensive code
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:13:31 +08:00
imacatandClaude Fable 5 8d223177cd Add the data layer with the build-db, fetch-lyrics, and fetch-artists subcommands
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 15:12:22 +08:00
imacatandClaude Fable 5 f9ab79f0c2 Use the unified pydantic-settings configuration in run_llm
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 14:03:37 +08:00
imacatandClaude Fable 5 b68393ab01 Record the data storage architecture and phase-1 plan in the project documents
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 14:03:37 +08:00
imacatandClaude Fable 5 12ace6f45a Add the tools sub-project with the run_llm batch runner
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-04 14:03:28 +08:00
imacatandClaude Fable 5 9a42b463a3 Initialize project structure, workflow rules, and research plan
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-30 18:51:10 +08:00
16 changed files with 142 additions and 495 deletions
+12 -12
View File
@@ -6,18 +6,18 @@
Messages API: model `claude-sonnet-4-6`, `temperature=0`,
thinking disabled, Batch API where possible.
- Prompt definition files live in
`prompts/<track>-<step>-<substep>-<task>.md` (e.g.
01-01-01-tag.md; the substep number fixes the execution
order; no version suffix -- versions live in git history)
and are passed verbatim as the system prompt.
- LLM steps whose outputs are item-by-item comparable
(convergence, coding) run the same definition file twice,
then a separate arbitration step settles only the
script-computed disagreements ("2 runs + 1 arbitration").
Free-generation steps run twice and both outputs are pooled,
unarbitrated. If arbitration output is unexpected, revise
the definition file and repeat that cycle; never patch
results by hand.
`prompts/<track>-<step>-<task>.md` (e.g. 01-01-tag.md; no
version suffix -- versions live in git history) and are
passed verbatim as the system prompt.
- Per-song LLM judgments (coding, screening) run the same
definition file twice, then a separate arbitration step
settles only the script-computed disagreements
("2 runs + 1 arbitration"). Free-generation steps run
twice and both outputs are pooled, unarbitrated.
Vocabulary-building steps (convergence) run once as a
recorded pass. If an arbitration or validation outcome is
unexpected, revise the definition file and repeat that
cycle; never patch results by hand.
- Each run of a step is archived self-contained under the
destination directory given explicitly on the `run-llm`
command line (by convention `runs/<definition-file>/run<N>/`):
+43 -50
View File
@@ -99,8 +99,8 @@
## 2026-08-02
- **`import-lyrics` 不設為子命令,pilot 歌詞改以 `excludes/`
私人腳本匯入**。理由:pilot 捕捉檔不隨論文發布,子命令
- **`import-lyrics` 不設為子命令,pilot 歌詞改以私人腳本
匯入**。理由:pilot 捕捉檔不隨論文發布,子命令
形式會在發布的 CLI 裡留下讀者無法執行的死命令——要交待的
是「沿用 pilot 捕捉」的事實(記於 lyrics-provenance.csv 與
論文方法節),不是工具本身;工具移出專案,發布管線即
@@ -178,8 +178,8 @@
來源公開可稽核,快照乾淨、overrides 縮小;論文方法節揭露
「捕捉前經研究者查證補完」。性別以公開自我認同為準,
推測不確定且無佐證者列疑慮清單;查證屬資料策展而非
分析,以 Claude Code 輔助、不走分析 APIQS 批次
`excludes/`私人工作檔,不隨論文發布
分析,以 Claude Code 輔助、不走分析 APIQS 批次
私人工作檔留存,不隨論文發布。
## 2026-08-04
@@ -227,7 +227,7 @@
- **Markdown 檔名一律以 dash 連接**decision-log.md、
research-plan.md、project-structure.md 等;與 data/ 層
CSV 檔名慣例一致),定義檔命名慣例同步改為
`prompts/<task>-v<N>.md`(如 screen-v1.md),`excludes/`
`prompts/<task>-v<N>.md`(如 screen-v1.md),版本庫外的
私人工作檔一併改名;`conference_abstract.md` 改名並搬入
`paper/`——它是本次年會實際送出的摘要,與全文同屬投稿
血脈,不是 `docs/` 的內部工作文件。全 repo 指涉同步更新。
@@ -323,51 +323,6 @@
式化、消滅隱性推導」原則的貫徹。比對子命令驗證兩 run 的定義檔與
輸入 SHA 一致、缺 run 即失敗。目標目錄已存在即拒絕執行,重跑須明
`--replace`
- **收斂仲裁演算法定案(逐對裁決)**:不做組對組匹配
(兩組是否「同一組」無原則性答案),比較化約為詞對的
共組關係——共識塊(交集細分)→ 分歧塊對枚舉 → 仲裁
逐對二元裁決(輸入匿名,不含執行別、共識與歌詞)→
union-find 遞移重組(固定邊集的連通元件唯一,與順序
無關,決定性成立)。仲裁自身亦 2+1:獨立執行兩次,
逐對比對,兩次相同即定案,相異者送 `-arb-arb` 終局票
(單次,依協定終局)——每個分歧對等同三票多數決,
仲裁鏈至此終止。邊界判斷之裁決為單次記錄性決定,不
宣稱可再生(可重現性依計畫定義=程序透明+可稽核)。
遞移導出的合併(共識「分開」的塊經第三塊橋接併組)為
等價關係語意的邏輯結論,照單接受並逐筆記錄供稽核。
一致率低於門檻(暫訂 50%,首輪實跑後校準)即不仲裁
——視為定義檔約束不足,修訂後重跑該循環。cap 仲裁後
組數略超 50 可接受,如實記錄。演算法明文於
`methodology.md`,供讀者依循重現。
- **收斂執行不命名,組名全權歸命名鏈(命名 2+1)**:
merge、cap 的執行輸出改為無名分組(陣列之陣列)——
既然命名鏈存在,收斂執行自取的名字毫無用處,徒增
兩次執行間「同組不同名」的假分歧。最終分組定案後,
`-name` 命名步驟獨立執行兩次對全部組自由命名——逐組
比對,同名即定案(單次執行內名字唯一,故定名間必無
撞名);異名送 `-name-arb` 擇一仲裁(匿名呈現兩候選,
輸出限於候選並避讓已定名,機械可驗證)。理由:組名
承重——merge 名為 cap 的輸入詞、cap 名為 code 的編碼
類目——凡承重的語意判斷皆受 2+1。cap 的輸入即 merge
定案後的組名清單,原始關鍵字不下傳——如同證據不進
收斂,上游細節不污染下游步驟。
- **定義檔輸出形狀統一與去署名**:LLM 輸出統一為單層
dictarray——tag、code:關鍵字→引述;merge、cap
無名分組(陣列之陣列);code-arb:保留關鍵字→仲裁者自己的引述
(剔除即不列,剔除集合由程式以鍵差推得);screen 系:
引述陣列,非空即「有」(present 布林刪除,判斷與依據
合一)。仲裁輸入一律不含執行別(assigned_by、
run1/run2 刪除)——標示何方主張會誘使仲裁者揣測
「哪次較可信」,而非就文本裁決。收斂組名採自由命名
(不限取自成員詞),取命名貼合度;可追溯性由軌跡歸檔
承擔。
- **定義檔改三層編號**
`prompts/<軌>-<步>-<次步>-<task>.md`——次步為步內執行
順序,明定讀者依循的先後(如 01-02-01-merge →
01-02-02-merge-arb → 01-02-03-merge-arb-arb →
01-02-04-merge-name → 01-02-05-merge-name-arb)。後綴
`-arb` 仲裁、`-arb-arb` 終局票、`-name` 命名、
`-name-arb` 命名擇一仲裁。
- **screen 標籤詞改為 `women-power`(來歷考據定案)**
考據先導研究的 local agent 存檔:其第一步指令含數十個
範例 thematic keywords,其中即有 women-power——為當時
@@ -381,3 +336,41 @@
考」敘述同步改為考據結果。附帶認清:先導第一步並非零
語意提示——此即正式研究「只定格式、不定語意」設計所
矯正者。
- **定義檔輸出形狀統一與去署名**:LLM 輸出統一為單層
dictarray——tag、code:關鍵字→引述;merge、cap
組名→成員;code-arb:保留關鍵字→仲裁者自己的引述
(剔除即不列,剔除集合由程式以鍵差推得);screen 系:
引述陣列,非空即「有」(present 布林刪除,判斷與依據
合一)。仲裁輸入一律不含執行別——標示何方主張會誘使
仲裁者揣測「哪次較可信」,而非就文本裁決。收斂組名採
自由命名(不限取自成員詞),取命名貼合度;可追溯性由
軌跡歸檔承擔。
- **定義檔改三層編號**
`prompts/<軌>-<步>-<次步>-<task>.md`——次步為步內執行
順序,明定讀者依循的先後(如 01-04-01-code →
01-04-02-code-arb);仲裁檔後綴 `-arb`
- **收斂演算法:檢視而棄用的方案**:詞彙表建構(merge、
cap)的重複執行與仲裁,歷經四個方案後全數棄用——
①整條管線獨立跑兩遍、於終點仲裁兩份最終詞彙表:兩套
分類系統互不可比,仲裁淪為第三次建構,一致率無從
定義。②組對組匹配(以相似度門檻判定兩組是否「同一
組」):無原則性答案,門檻任意。③tag 步驟仲裁:自由
詞彙兩次輸出不共享比對單位,無物可裁(tag 改為兩次
進池,沿用至今)。④逐對仲裁鏈:交集細分出共識塊、
分歧塊對三票多數決、union-find 遞移重組、仲裁後命名
2+1——機械上可行且逐項可驗(共 15 份定義檔,全版
保存於分支 `tag-algo-13` 備考),但其變異縮減未經證實
(逐對多數決降低對層變異,遞移閉包卻放大結構層變異,
淨效果不明),複雜度成本則屬確定,且縮減的是儀器變異
——對量測無關緊要的量(見次條)。
- **詞彙表建構改為單次記錄性程序(演算法簡化)**:
merge、cap 各單次執行、執行內自行命名(具名分組
輸出),定義檔減為 7 份。2+1 原則改寫:語料層逐首
判斷(code、screen)一律 2+1;自由生成(tag)兩次
進池;詞彙表建構單次、全程歸檔。理由:(1) 影響量測的
是編碼層——詞彙表屬揭露的儀器選擇,凍結後下游同尺量
到底,其抽樣變異不污染量測;(2) 對齊領域慣行——
codebook 建構本為單次詮釋程序,信度檢驗施於編碼應用
層;(3) 複雜仲裁機械無以自證其益(見前條)。驗證改為
確定性格式檢查(完整分割、組名唯一、cap ≤ 50),違規
依協定修訂定義檔重跑。
+40 -114
View File
@@ -6,105 +6,33 @@
## 自然編碼管線總覽
四步驟:自由標註(tag)→ 自然收斂(merge)→ 強制收斂
cap)→ 全量編碼(code),另設「女性力量」單目標篩選
screen)作黃金標準取樣的補漏網。歌詞只出現在 tag、code、
screen 與仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
歌詞。設計原則見 `research-plan.md`本檔記載可重現的
演算法細節。
四步驟:自由標註(tag,兩次進池)→ 自然收斂(merge
單次)→ 強制收斂cap,單次)→ 全量編碼(code2+1),
另設「女性力量」單目標篩選screen2+1)作黃金標準取樣
的補漏網。歌詞只出現在 tag、code、screen 與仲裁步驟;
merge、cap 不接觸歌詞。設計原則見 `research-plan.md`
本檔記載可重現的演算法細節。
## 收斂步驟merge、cap的 2+1 比對與仲裁
## 詞彙表建構merge、cap——單次記錄性程序
兩次獨立執行對同一批輸入詞各產生一個分組。比對不做
「組對組」的匹配——兩個組是否為「同一組的變體」無原則性
答案——而是把比較化約為「詞對的共組關係」:
merge 與 cap 各以單次執行完成,執行內自行為各組命名,
輸出具名分組(組名 → 成員詞)。不做重複執行與仲裁。
1. **共識塊(交集細分)**:兩次執行都放在同組的詞歸為
同一塊。即以「(第一次的組, 第二次的組)」二元組為鍵
分桶,一桶一塊。此步為純集合運算。
2. **分歧塊對枚舉**:每個塊完整落在各次執行的恰一組內,
故「兩塊在某次執行中是否同組」定義良好。逐塊對檢查:
兩次執行答案相同者為共識(同組或分開,直接定案);
不同者列入分歧清單
3. **逐對仲裁(自身 2+1**:分歧塊對送 LLM 仲裁
`01-02-02-merge-arb.md``01-03-02-cap-arb.md`)。
仲裁者只看兩塊的內容詞,逐對二元裁決「是否同一
主題」;輸入不含執行別(避免「猜哪一次較可信」的
偏誤),不含共識部份,不含歌詞。仲裁自身獨立執行
兩次,逐對比對:兩次裁決相同即定案;相異的塊對送
終局票(`01-02-03-merge-arb-arb.md`
`01-03-03-cap-arb-arb.md`,單次執行,依協定為終局)
——每個分歧塊對等同三票多數決,仲裁鏈至此終止。
送入仲裁的塊對即模型自身不穩定的邊界判斷,其裁決為
單次記錄性決定,不宣稱可再生;可重現性依計畫定義為
「程序透明+可稽核」,裁決與其輸入全程歸檔。
4. **確定性重組**:以每個「同組」裁決為一條邊,最終
分組=圖的連通元件(union-find)。固定邊集的連通
元件唯一,與處理順序無關,故重組決定性成立。
遞移性後果照單全收:兩個共識「分開」的塊可能經第三
塊橋接而併入同組——此為等價關係語意的邏輯結論,
比對程式將此類「遞移導出的合併」逐筆記錄於歸檔,
供稽核。仲裁的結果空間因此大於「兩次執行擇一」:
可能比兩次都粗(多對皆裁同組),也可能比兩次都細
(多對皆裁分開)。
5. **一致率與保險絲**:比對程式計算塊對層級的一致率並
記入歸檔。低於門檻(暫訂 50%,首輪實跑後校準)即
不進行仲裁——兩次分組面目全非說明定義檔約束不足,
依協定修訂定義檔並重跑該循環。
6. **cap 的上限容忍**:仲裁後組數可能略超 50(多對裁
「分開」時)。略微超過可接受,如實記錄,不強行
壓縮。
7. **組名定案(收斂不命名+命名 2+1)**merge、cap 的
執行輸出不含組名(無名分組,陣列之陣列)——命名鏈
既然存在,收斂執行自取的名字只會製造兩次執行間
「同組不同名」的假分歧。組名承重——merge 的組名是
cap 的輸入詞,cap 的組名是 code 對歌詞編碼的類目——
故與其他語意判斷同受 2+1。程序:(1) 最終分組定案後,
命名步驟(`01-02-04-merge-name.md`
`01-03-04-cap-name.md`)獨立執行兩次,輸入為不透明
組 ID 對成員詞,對全部組自由命名,僅受格式約束
(小寫連字號、單次執行內不重複);(2) 逐組比對:兩次
同名即定案(單次執行內名字唯一,故定名間必無撞名);
(3) 異名的組送擇一仲裁(`01-02-05-merge-name-arb.md`
`01-03-05-cap-name-arb.md`)——匿名呈現兩候選,逐組
擇一,輸出限於候選並避讓已定名,機械可驗證。cap 的
輸入即 merge 定案後的組名清單。
### 收斂步驟的資料流(比對子命令的輸入輸出契約)
以 merge 為例(cap 完全同構,檔名換為 01-03 系):
1. 兩次執行的原始輸出:`runs/01-02-01-merge/run1/output.jsonl`
`run2/output.jsonl`,各含一個無名分組(陣列之陣列)。
2. 比對子命令讀入兩份分組,先驗證兩者為同一輸入詞集的
完整分割(缺詞、多詞、重複即失敗),再計算共識塊與
分歧塊對,產出仲裁輸入檔——即
`01-02-02-merge-arb.md` 所收的 JSON
- `blocks`:塊 ID → 成員詞。**塊 ID 的指派決定性**:
全部塊先按「各塊字典序最小的成員詞」排序,依序編為
b1、b2、…;塊內成員詞亦按字典序排列。
- `pairs`:分歧塊對清單,每對內部按塊 ID 序、清單
整體按 (第一元素, 第二元素) 字典序排列。
- 僅分歧塊對入列;共識(同組或分開)不送仲裁,由
比對子命令直接寫入共識紀錄檔。
同時產出:共識紀錄(共識塊、共識同組對、共識分開對)
與塊對一致率(含保險絲判定)。
3. 仲裁輸入檔以 run-llm 跑兩次
`runs/01-02-02-merge-arb/run1``run2`);比對
子命令逐對比對兩份裁決,兩票相同即定案,相異的塊對
依同一契約組成終局票輸入檔(`blocks` 僅含涉事塊、
ID 沿用原編號),跑
`runs/01-02-03-merge-arb-arb/run1`
4. 裁決套用子命令彙整三票結果,以 union-find 重組出
最終分組,並寫出:最終分組檔(無名,陣列之陣列,
組序與組內成員皆字典序)、遞移導出合併的紀錄、
收斂軌跡(原始詞 → 最終組)。
5. 最終分組轉為不透明組 ID(依組序編 g1、g2、…)進入
命名鏈(見第 7 條);命名定案後,組名清單(字典序)
即下一步的輸入。
6. 以上中間檔全部隨 `runs/` 歸檔;一切排序規則固定,
故給定相同的兩份執行輸出與相同的裁決,全流程輸出
逐位元組可再生。
- **定位**:詞彙表是研究的儀器,不是量測。詞彙表凍結
入庫後,下游全部量測(編碼 vs 黃金標準)以同一把尺
進行,內部一致;建構過程的抽樣變異屬「揭露的儀器
選擇」,不污染量測。此與質性研究慣行一致——codebook
建構本為單次的詮釋程序,信度檢驗施於編碼應用層。
- **取捨紀錄**:曾設計逐對仲裁鏈(共識塊、分歧塊對三票
多數、命名 2+1;完整版本保存於分支 `tag-algo-13`
棄用理由:逐對多數決雖降低單對變異,但遞移閉包會放大
結構層變異,淨縮減未經證實;其複雜度成本卻是確定的;
且縮減的是儀器變異——對論文主張無關緊要的量。
- **確定性驗證**(違規即依協定修訂定義檔重跑):輸出須
為輸入詞集的完整分割(缺詞、多詞、重複即失敗);組名
唯一且符合格式;cap 組數 ≤ 50。
- **收斂軌跡**:原始關鍵字 → merge 組名 → cap 組名,
兩份執行輸出本身即完整記錄。
## 編碼步驟(code、screen)的 2+1 比對與仲裁
@@ -120,6 +48,9 @@ screen 與各仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
引述陣列。
- 仲裁者的引述可能與原引述不同:仲裁是對歌詞的重新
判讀,其引述是該裁決自身的依據,非轉抄。
- 送入仲裁的標籤即模型自身不穩定的邊界判斷,其裁決為
單次記錄性決定,不宣稱可再生;可重現性依計畫定義為
「程序透明+可稽核」,裁決與其輸入全程歸檔。
## 軌跡對映(診斷用)
@@ -130,7 +61,7 @@ screen 與各仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
## 全管線的交接契約
每一步的輸出如何變成下一步的輸入,皆為確定性程序,規則
明定如下(收斂步驟內部的交接見上節)
明定如下:
- **歌詞輸入檔(tag、code、screen 共用)**
`export-llm-input` 自工作儲存產出,每筆
@@ -142,9 +73,9 @@ screen 與各仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
進池同時寫出處記錄(關鍵字 →(執行別,歌曲 ID)
清單),供軌跡對映回到歌曲;出處記錄不進任何 LLM
輸入。
- **merge 定案 → cap**merge 定案組名以字典序排成
JSON 陣列,即 cap 的輸入(見上節第 5 點)
- **cap 定案 → code 定義檔**定案組名以字典序逐行填入
- **merge → cap**merge 輸出的組名以字典序排成 JSON
陣列,即 cap 的輸入;原始關鍵字不下傳
- **cap → code 定義檔**cap 輸出的組名以字典序逐行填入
`01-04-01-code.md` 的詞彙表節(逐字),檔案隨 git
commit 後方可執行——code 的定義檔因此自我完備,
論文附錄可直接引用。
@@ -161,11 +92,6 @@ screen 與各仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
入仲裁輸入 JSONL(依 ID 升序),`content`
`{"lyrics": …, "evidence": <肯定方引述>}`
- **screen 定案**:命中集合=兩次皆有 ∪ 仲裁裁定有。
- **命名鏈的輸入構成**`-name` 輸入的組 ID 依最終分組
之組序(上節第 5 點)編 g1、g2、…,組內成員字典序;
`-name-arb` 輸入中每組的兩個候選名**按字典序排列**
——不按執行別,避免順序洩漏何方所取;`taken` 為已
定案名的字典序清單。
- **女性力量候選集**:於 cap 詞彙表定案後、黃金標準
編碼開始前,由研究者指認詞彙表中屬「女性力量」概念
域的組(指認及理由記入決策日誌),候選集=code 定案
@@ -173,20 +99,20 @@ screen 與各仲裁步驟;merge、cap 及其仲裁、命名皆不接觸
- **序列化通則**:所有中間檔為 UTF-8 JSON,鍵序與元素
序皆依上列規則明定,無時間戳、無隨機成分;解析一律
偵測重複鍵,違規即失敗。JSONL 一律以換行字元(\n)
切行——歌詞含 U+2028 等 Unicode 行分隔符
切行——歌詞含 U+0085 等控制字元時
`str.splitlines()` 類的通用切行會截斷 JSON 字串
(實測踩中)。給定相同的 LLM 執行輸出,
全部交接產物逐位元組可再生。
(實測踩中)。給定相同的 LLM 執行輸出,全部交接產物
逐位元組可再生。
## 執行與稽核
- 每一步驟以 `run-llm <定義檔> <輸入檔> <歸檔目錄>`
執行;獨立執行兩次=重現命令清單上的兩行命令,各自
歸檔(`runs/<定義檔名>/run1``run2`),仲裁與命名
各為獨立步驟、獨立歸檔
- 比對、裁決套用、重組、對映皆為確定性程式(子命令),
其輸入輸出檔隨 runs/ 歸檔,JSON 解析一律偵測重複鍵
違規即失敗
執行;2+1 步驟的兩次執行=重現命令清單上的兩行命令,
各自歸檔(`runs/<定義檔名>/run1``run2`),仲裁
獨立步驟、獨立歸檔;單次步驟(merge、cap)一行命令、
一份歸檔。
- 進池、比對、裁決套用、對映皆為確定性程式(子命令)
其輸入輸出檔隨 runs/ 歸檔
- Batch API 的每筆請求自含全部脈絡且互不可見(平台
契約),歌與歌之間的獨立性由此成立;兩次執行的獨立
性由「兩次呼叫、兩個批次、兩份歸檔」的執行結構自明。
+2 -2
View File
@@ -25,7 +25,7 @@ pop-fem-audit/
│ └── artists.csv # 歌手報表(人讀;進 git)
├── prompts/ # LLM 定義檔(逐字作為 system prompt
│ └── <軌>-<步>-<次步>-<task>.md # 01-01-01-tag.md、
│ # 01-02-02-merge-arb.md…
│ # 01-04-02-code-arb.md…
│ # 不帶版本號,版本即 git 歷史
├── tools/ # 輔助工具子專案(src-layout
│ ├── pyproject.toml # 發行名 pop-fem-audit-tools
@@ -73,7 +73,7 @@ pop-fem-audit/
│ └── methodology.md # 方法細節(全文方法節底稿;
│ # 映射分析方法須在看結果前寫定)
└── paper/
├── conference-abstract.md # pilot 摘要(投稿版)
├── abstract.md # 摘要
└── full-paper.md # 全文
```
+27 -31
View File
@@ -7,17 +7,20 @@
- **研究敘事**:全文為獨立完整的正式研究(2016–2025 十年語料
共 1000 筆榜單紀錄,凍結程序後全量執行),**不與先導研究
paper/conference-abstract.md做比較**;全文數字一律以正式研究
做比較**;全文數字一律以正式研究
結果為準。先導研究僅作為 codebook v0 與假說的內部來源,
記於決策日誌,不進入論文敘事。
- **執行原則**:主會話只做討論;所有分析由 deterministic script
執行。LLM 步驟以 Python script 呼叫 Anthropic Messages API
(個人 Console 帳號、Batch API 五折),定義檔逐字作為 system
prompt。2+1 協定依輸出可比性適用:輸出可逐項機械比對的步驟
(收斂、編碼)「同一定義檔獨立執行兩次+一次仲裁」,仲裁
只裁程式算出的分歧清單;自由生成步驟(首步自由標註)兩次
執行全數進池、不仲裁——自由詞彙兩次輸出不共享比對單位,
無物可裁。仲裁結果不符預期則修訂定義檔重跑該循環。
prompt。2+1 協定適用於語料層的逐首判斷(編碼、篩選):
「同一定義檔獨立執行兩次+一次仲裁」,仲裁只裁程式算出
的分歧。自由生成步驟(自由標註)兩次執行全數進池、不
仲裁——自由詞彙兩次輸出不共享比對單位,無物可裁。詞彙
表建構(收斂)為單次記錄性程序,不重複執行——詞彙表是
揭露的儀器選擇,非量測;信度檢驗施於編碼層(詳見
`methodology.md`)。仲裁或驗證結果不符預期則修訂定義檔
重跑該循環。
- **提示詞只定格式、不定語意**:研究對象是通用 LLM 以其
網路語料知識背景所做的自然編碼,編碼結果本身是批判對象。
LLM 定義檔只規定任務形狀(輸入、數量範圍、輸出格式),
@@ -69,8 +72,8 @@
候選歌曲的歌手另做深度背景(族裔以公開自我認同為準、音樂
場景),script 輔助、人工核定。
- **Pilot 歌詞沿用(私人匯入,不進發布管線)**:先導研究
捕捉檔(lyrics.json684 首,20182025)以 `excludes/`
私人腳本匯入歌詞快取——只取識別欄位與歌詞本文,以
捕捉檔(lyrics.json684 首,20182025)以私人腳本
匯入歌詞快取——只取識別欄位與歌詞本文,以
(year, rank) 精確匹配 song_id。匯入工具不屬於專案交付物
(讀者拿不到其輸入),讀者的重現路徑純粹是 `fetch-lyrics`
沿用之**事實**記於 `data/captures/lyrics-provenance.csv`(進 git):
@@ -95,13 +98,12 @@
themes 過早抽象)之繼承,於執行前鎖定,防止事後擇優。
2. **自然收斂(merge**:輸入為池中純去重關鍵字清單——
無歌詞、無頻次、無歌曲出處——模型按自身理解合併近似
概念。2+1:兩次收斂處理同一批輸入詞,程式算出共識塊
與分歧塊對,逐對仲裁(仲裁自身亦 2+1,第三票終局)後
以連通元件決定性重組;收斂執行不命名,組名由命名步驟
2+1 定案(演算法詳見 `methodology.md`)。
概念,執行內自行為各組命名。單次執行,不仲裁;確定性
驗證(完整分割、組名唯一)違規即修訂重跑(詳見
`methodology.md`)。
3. **強制收斂(cap**:同第 2 步形態,輸入為 merge
定案後的組名清單(原始關鍵字不下傳),限制併至 50 個
以內;仲裁後略超 50 可接受,如實記錄。頻次不入收斂:頻率的分析角色由第 4 步編碼承擔;
輸出的組名清單(原始關鍵字不下傳),限制併至 50 個
以內。單次執行,不仲裁。頻次不入收斂:頻率的分析角色由第 4 步編碼承擔;
池中頻次含跨執行噪音;頻次會誘使模型以頻率剪枝(喪失
稀有主題)並把高頻大主題切細。代價(特異主題佔名額)
已知並接受,換取主題多樣性與純語意歸併的可辯護性。
@@ -123,18 +125,16 @@
提示中唯一的語意種子,屬研究者的儀器選擇,據實揭露。
定義檔命名 `prompts/<軌>-<步>-<次步>-<task>.md`——軌
01=由下而上自然編碼、02=預先決定的 women-power
篩選;步為軌內步驟序,次步為
步內執行順序,讀者依編號先後依循(如 01-02-01-merge.md →
01-02-02-merge-arb.md → 01-02-03-merge-arb-arb.md
01-02-04-merge-name.md → 01-02-05-merge-name-arb.md)。
後綴 `-arb` 為仲裁、`-arb-arb` 為仲裁之終局票、`-name`
為命名、`-name-arb` 為命名擇一仲裁。檔名不帶
版本號——版本即 git 歷史,失敗的版本不保留,需要回看的
舊版都在 git history;每次執行的定義檔快照隨 `runs/`
自我完備。收斂執行的輸出本身即分組記錄(無名分組
JSON),與全部中間交接檔同隨 `runs/` 歸檔(交接契約見
`methodology.md`)。
01=由下而上自然編碼、02=預先決定的 women-power 篩選;
步為軌內步驟序,次步為步內執行順序,讀者依編號先後依循:
01-01-01-tag.md、01-02-01-merge.md、01-03-01-cap.md、
01-04-01-code.md、01-04-02-code-arb.md
02-01-01-screen.md、02-01-02-screen-arb.md;仲裁定義檔
同 prefix 加 `-arb`。檔名不帶版本號——版本即 git 歷史,
失敗的版本不保留,需要回看的舊版都在 git history;每次
執行的定義檔快照隨 `runs/` 自我完備。收斂執行的輸出本身
即分組記錄(具名分組 JSON),與全部中間交接檔同隨
`runs/` 歸檔(交接契約見 `methodology.md`)。
## 階段與時程(全文截稿 2026-08-15
@@ -142,7 +142,7 @@ JSON),與全部中間交接檔同隨 `runs/` 歸檔(交接契約見
|---|---|---|---|
| 0 | 基礎建設:git init、目錄結構、.gitignore、決策日誌、runner script(含 Batch API)、codebook v0 骨架 | script + 討論 | 7/307/31 |
| 1 | 資料準備:`run_llm` 改走統一設定 → `build-db`(解析榜單成 songs/chart_entries/artists/song_artists)→ `import-lyrics`pilot 20182025)→ `fetch-lyrics`201617 與缺漏,Lyrics.ovh / LRCLIB)→ `fetch-artists`Wikidata 快照)→ `export-llm-input` | 子命令 | 7/318/3 |
| 2 | 自然編碼管線:tag ×2 進池 → merge 2+1 → cap 2+1 → 詞彙表定稿 → code 2+1(全 883 首,附引述);另跑 screen 補漏網 | API + script | 8/48/7 |
| 2 | 自然編碼管線:tag ×2 進池 → merge ×1 → cap ×1 → 詞彙表定稿 → code 2+1(全 883 首,附引述);另跑 screen 補漏網 | API + script | 8/48/7 |
| 3 | 黃金標準:依 codebook 人工逐首判定 genuine/peripheral/fake,附引用歌詞證據表(LLM 只做摘錄,不給判定建議);先以 10–15 首校準樣本試編並修訂 codebook 後凍結;同批校準樣本實測 Sonnet 4.6 vs Opus 5 一致率 | 人工 + script 輔助 | 8/58/9 |
| 4 | 受控比較(盲點實驗):條件 A(詞彙層提示)vs 條件 B(框架感知提示),各 2+1,對照黃金標準計算假陽/假陰率 | API | 8/88/11 |
| 4' | 映射分析:自然編碼結果(第 4 步)與黃金標準交叉表;軌跡對映 vs 直接編碼的扭曲診斷(分析方法先寫入 methodology.md 再看結果) | script | 與 4 並行 |
@@ -165,14 +165,10 @@ JSON),與全部中間交接檔同隨 `runs/` 歸檔(交接契約見
與「貶低其他女性/以男性指標定義權力」。
- 流程:v0 → 校準樣本試編 → 修訂 → 凍結 v1 → 全量編碼;
中途修訂須記決策日誌並回檢已編歌曲。
- Positionality statement 寫入論文(53 歲、長年婦運者、資深
工程師、離開學術圈多年、因整理榜單而起)。
## 其他已定案事項
- 歌詞受版權保護:完整歌詞不進 git(`data/captures/lyrics/` gitignored),
論文與 repo 只留分類所引摘錄。
- 工程支援(寫 script、style check、審稿)用公司訂閱的
Claude Code;進論文的分析 token 由個人 API 帳號支付。
- Fable 5 不用於 pipeline:成本高、thinking 無法關閉且不可稽核、
無 temperature 控制,且會混淆「盲點是提示問題」的核心主張。
+9 -6
View File
@@ -13,11 +13,14 @@ Rules:
- Do not invent keywords that are not in the input.
- Merge only what you consider the same theme; a group may be
a single keyword.
- Name each group with a short lowercase phrase, words joined
by hyphens, that best names the theme it gathers. Group
names must be unique.
Output a single JSON array of groups — each group an array of
its member keywords and nothing else:
Output a single JSON object mapping each group name to the
list of its member keywords, and nothing else:
[
["keyword", "another-keyword"],
["keyword"]
]
{
"first-group-name": ["member-keyword", "another-member"],
"second-group-name": ["member-keyword"]
}
-36
View File
@@ -1,36 +0,0 @@
You are arbitrating between two independent consolidations of
the same vocabulary of thematic keywords from a song corpus.
Where the two consolidations agree, the result has been
settled by script and is not shown. You rule on the
disagreements: pairs of keyword blocks that one consolidation
merged into one theme and the other kept apart.
Input: a JSON object. "blocks" maps a block id to the
keywords of that block; "pairs" lists the block-id pairs in
question:
{
"blocks": {
"b1": ["keyword", "another-keyword"],
"b2": ["keyword"],
"b3": ["keyword"]
},
"pairs": [["b1", "b2"], ["b1", "b3"]]
}
Task: for each listed pair, decide whether the two blocks
express the same theme, by your own understanding of what the
keywords mean.
Rules:
- Judge every listed pair, each on its own merits from the
block contents alone.
- Judge only the listed pairs.
Output a single JSON array holding the pairs whose two blocks
express the same theme, written exactly as given in "pairs";
an empty array when none do:
[["b1", "b2"]]
-34
View File
@@ -1,34 +0,0 @@
You are casting the deciding judgment on disagreements about
a consolidated vocabulary of thematic keywords from a song
corpus: for each pair of keyword blocks in question, earlier
independent judgments disagreed on whether the two blocks
express the same theme.
Input: a JSON object. "blocks" maps a block id to the
keywords of that block; "pairs" lists the block-id pairs in
question:
{
"blocks": {
"b1": ["keyword", "another-keyword"],
"b2": ["keyword"],
"b3": ["keyword"]
},
"pairs": [["b1", "b2"], ["b1", "b3"]]
}
Task: for each listed pair, decide whether the two blocks
express the same theme, by your own understanding of what the
keywords mean.
Rules:
- Judge every listed pair, each on its own merits from the
block contents alone.
- Judge only the listed pairs.
Output a single JSON array holding the pairs whose two blocks
express the same theme, written exactly as given in "pairs";
an empty array when none do:
[["b1", "b2"]]
-28
View File
@@ -1,28 +0,0 @@
You are naming the theme groups of a consolidated vocabulary
of thematic keywords from a song corpus.
Input: a JSON object mapping an opaque group id to the member
keywords of that group:
{
"g1": ["keyword", "another-keyword"],
"g2": ["keyword"]
}
Task: give each group a name that best names the theme its
members gather, by your own understanding of what the
keywords mean.
Rules:
- Each name is a short lowercase phrase with the words joined
by hyphens.
- Names must be unique across the groups.
Output a single JSON object mapping each group id to its
name, and nothing else:
{
"g1": "group-name",
"g2": "another-group-name"
}
-37
View File
@@ -1,37 +0,0 @@
You are arbitrating between two independent namings of the
theme groups of a consolidated vocabulary of thematic
keywords from a song corpus.
For each group in question, the two namings proposed
different names. You choose between them.
Input: a JSON object. "groups" maps an opaque group id to
its member keywords and its two candidate names; "taken"
lists names that are already in use:
{
"groups": {
"g1": {
"members": ["keyword", "another-keyword"],
"candidates": ["one-name", "other-name"]
}
},
"taken": ["existing-name"]
}
Task: for each group, choose the candidate that better names
the theme its members gather, by your own understanding of
what the keywords mean.
Rules:
- Choose only from that group's two candidates.
- Choices must be unique across the groups and must not
reuse any name in "taken".
Output a single JSON object mapping each group id to the
chosen name, and nothing else:
{
"g1": "one-name"
}
+9 -6
View File
@@ -16,11 +16,14 @@ Rules:
- Merge the keywords that are closest in meaning first; keep
clearly distinct themes apart as far as the limit allows.
A group may be a single keyword.
- Name each group with a short lowercase phrase, words joined
by hyphens, that best names the theme it gathers. Group
names must be unique.
Output a single JSON array of groups — each group an array of
its member keywords and nothing else:
Output a single JSON object mapping each group name to the
list of its member keywords, and nothing else:
[
["keyword", "another-keyword"],
["keyword"]
]
{
"first-group-name": ["member-keyword", "another-member"],
"second-group-name": ["member-keyword"]
}
-37
View File
@@ -1,37 +0,0 @@
You are arbitrating between two independent consolidations
that each reduced the same vocabulary of thematic keywords
from a song corpus to a fixed maximum number of themes.
Where the two consolidations agree, the result has been
settled by script and is not shown. You rule on the
disagreements: pairs of keyword blocks that one consolidation
merged into one theme and the other kept apart.
Input: a JSON object. "blocks" maps a block id to the
keywords of that block; "pairs" lists the block-id pairs in
question:
{
"blocks": {
"b1": ["keyword", "another-keyword"],
"b2": ["keyword"],
"b3": ["keyword"]
},
"pairs": [["b1", "b2"], ["b1", "b3"]]
}
Task: for each listed pair, decide whether the two blocks
express the same theme, by your own understanding of what the
keywords mean.
Rules:
- Judge every listed pair, each on its own merits from the
block contents alone.
- Judge only the listed pairs.
Output a single JSON array holding the pairs whose two blocks
express the same theme, written exactly as given in "pairs";
an empty array when none do:
[["b1", "b2"]]
-35
View File
@@ -1,35 +0,0 @@
You are casting the deciding judgment on disagreements about
a vocabulary of thematic keywords from a song corpus that was
consolidated to a fixed maximum number of themes: for each
pair of keyword blocks in question, earlier independent
judgments disagreed on whether the two blocks express the
same theme.
Input: a JSON object. "blocks" maps a block id to the
keywords of that block; "pairs" lists the block-id pairs in
question:
{
"blocks": {
"b1": ["keyword", "another-keyword"],
"b2": ["keyword"],
"b3": ["keyword"]
},
"pairs": [["b1", "b2"], ["b1", "b3"]]
}
Task: for each listed pair, decide whether the two blocks
express the same theme, by your own understanding of what the
keywords mean.
Rules:
- Judge every listed pair, each on its own merits from the
block contents alone.
- Judge only the listed pairs.
Output a single JSON array holding the pairs whose two blocks
express the same theme, written exactly as given in "pairs";
an empty array when none do:
[["b1", "b2"]]
-29
View File
@@ -1,29 +0,0 @@
You are naming the theme groups of a vocabulary of thematic
keywords from a song corpus that was consolidated to a fixed
maximum number of themes.
Input: a JSON object mapping an opaque group id to the member
keywords of that group:
{
"g1": ["keyword", "another-keyword"],
"g2": ["keyword"]
}
Task: give each group a name that best names the theme its
members gather, by your own understanding of what the
keywords mean.
Rules:
- Each name is a short lowercase phrase with the words joined
by hyphens.
- Names must be unique across the groups.
Output a single JSON object mapping each group id to its
name, and nothing else:
{
"g1": "group-name",
"g2": "another-group-name"
}
-38
View File
@@ -1,38 +0,0 @@
You are arbitrating between two independent namings of the
theme groups of a vocabulary of thematic keywords from a song
corpus that was consolidated to a fixed maximum number of
themes.
For each group in question, the two namings proposed
different names. You choose between them.
Input: a JSON object. "groups" maps an opaque group id to
its member keywords and its two candidate names; "taken"
lists names that are already in use:
{
"groups": {
"g1": {
"members": ["keyword", "another-keyword"],
"candidates": ["one-name", "other-name"]
}
},
"taken": ["existing-name"]
}
Task: for each group, choose the candidate that better names
the theme its members gather, by your own understanding of
what the keywords mean.
Rules:
- Choose only from that group's two candidates.
- Choices must be unique across the groups and must not
reuse any name in "taken".
Output a single JSON object mapping each group id to the
chosen name, and nothing else:
{
"g1": "one-name"
}