diff --git a/agent/tests/factors/test_alpha_library_translations.py b/agent/tests/factors/test_alpha_library_translations.py
new file mode 100644
index 00000000..10598fcf
--- /dev/null
+++ b/agent/tests/factors/test_alpha_library_translations.py
@@ -0,0 +1,52 @@
+"""The Alpha Library's Chinese notes, nicknames and theme names match the registry.
+
+The wiki shows every alpha's ``notes`` and ``nickname`` in Chinese from
+``wiki/alpha-library/i18n.zh.json``. Each entry keeps the English it was
+translated from, and the page generator only uses a translation whose English
+still equals the registry's, so an edited note silently falls back to English.
+These tests make that visible: a new alpha with a note but no translation, an
+edited note whose translation is now stale, a translation for an alpha that no
+longer exists, or a theme tag with no Chinese name all fail here.
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+
+import pytest
+
+from src.factors.registry import Registry
+
+STORE = Path(__file__).resolve().parents[3] / "wiki" / "alpha-library" / "i18n.zh.json"
+
+
+@pytest.fixture(scope="module")
+def registry_meta() -> dict[str, dict]:
+    manifest = Registry().export_manifest()
+    return {a["id"]: a["meta"] for zoo in manifest["zoos"] for a in zoo["alphas"]}
+
+
+@pytest.fixture(scope="module")
+def store() -> dict:
+    return json.loads(STORE.read_text(encoding="utf-8"))
+
+
+@pytest.mark.parametrize(("field", "bucket"), [("notes", "notes"), ("nickname", "nicknames")])
+def test_every_english_text_has_a_current_chinese_version(
+    registry_meta: dict[str, dict], store: dict, field: str, bucket: str
+) -> None:
+    english = {aid: meta[field] for aid, meta in registry_meta.items() if (meta.get(field) or "").strip()}
+    translated = store[bucket]
+
+    assert not sorted(set(english) - set(translated)), "untranslated: add them to i18n.zh.json"
+    assert not sorted(set(translated) - set(english)), "translations for alphas that have no such text any more"
+    stale = sorted(aid for aid, text in english.items() if translated[aid]["en"] != text)
+    assert not stale, f"English changed since translation, update 'en' and 'zh': {stale}"
+    assert all(entry["zh"].strip() and entry["zh"] != entry["en"] for entry in translated.values())
+
+
+def test_every_theme_tag_has_a_chinese_name(registry_meta: dict[str, dict], store: dict) -> None:
+    used = {tag for meta in registry_meta.values() for tag in meta.get("theme") or []}
+
+    assert used <= set(store["themes"]), sorted(used - set(store["themes"]))
diff --git a/wiki/alpha-library/alpha.js b/wiki/alpha-library/alpha.js
index 763a1092..cc7fe751 100644
--- a/wiki/alpha-library/alpha.js
+++ b/wiki/alpha-library/alpha.js
@@ -1,6 +1,6 @@
 // Fills the Alpha Library landing stats from content/index.json, which the
 // deploy workflow generates (wiki/scripts/build_alpha_library.py).
-import { storedLang } from "/site.js?v=20260929d";
+import { storedLang } from "/site.js?v=20260929e";
 
 const TEXT = {
   en: { across: (n) => `across ${n} zoos`, alphas: (n) => `${n} alphas`, missing: "manifest not generated yet" },
diff --git a/wiki/alpha-library/i18n.zh.json b/wiki/alpha-library/i18n.zh.json
new file mode 100644
index 00000000..0bbdd486
--- /dev/null
+++ b/wiki/alpha-library/i18n.zh.json
@@ -0,0 +1,1250 @@
+{
+ "notes": {
+  "academic_bab": {
+   "en": "Frazzini & Pedersen (2014) BAB. Price-only proxy: market return is the cross-sectional mean of daily per-stock returns (equal-weighted, not value-weighted as in the original). Beta is Cov(r_i, r_m) / Var(r_m) over a single 252-day window (paper uses separate 1y vol / 5y correlation windows). Factor score is negative beta, cross-sectional z-scored so low-beta names rank highest, matching the long leg of the original long-short construction.",
+   "zh": "Frazzini & Pedersen (2014) 低贝塔策略（BAB）。纯价格代理：市场收益率取每日个股收益率的截面均值（等权，而非原文的市值加权）。贝塔为单一 252 日窗口上的 Cov(r_i, r_m) / Var(r_m)（论文分别使用 1y 波动率窗口与 5y 相关系数窗口）。因子得分为负贝塔，经截面 z 分数标准化，使低贝塔标的排名最高，与原文多空构造中的多头腿一致。"
+  },
+  "academic_carhart_mom": {
+   "en": "Carhart (1997) UMD momentum factor. 12-month return minus 1-month return, cross-sectional z-score per date for long-short ranking. Top z-scores = winners. Constructed directly from prices, so this matches the original definition modulo the z-score wrapper. Canonical 252d window; declared decay_horizon=60 due to registry schema cap (le=60); real signal horizon=252.",
+   "zh": "Carhart (1997) UMD 动量因子。12 个月收益率减去 1 个月收益率，按日期做截面 z 分数，用于多空排名。z 分数最高 = 赢家。直接由价格构造，因此除外层的 z 分数变换外与原始定义一致。采用标准的 252d 窗口；受注册表 schema 上限（le=60）所限，声明 decay_horizon=60；实际信号周期=252。"
+  },
+  "academic_cma": {
+   "en": "[PRICE PROXY] for the Fama-French (2015) CMA (Conservative Minus Aggressive) investment factor. The original definition uses total-asset growth from fundamental data; here we use the negative 60-day change in log average volume as an activity-growth proxy, then cross-sectional z-score per date for long-short ranking. Top z-scores = volume contraction (conservative / low-growth proxy).",
+   "zh": "[价格代理] 对应 Fama-French (2015) CMA（Conservative Minus Aggressive，保守减激进）投资因子。原始定义使用基本面数据中的总资产增长；此处改用对数平均成交量 60 日变化的负值作为活跃度增长代理，再按日期做截面 z 分数，用于多空排名。z 分数最高 = 成交量收缩（保守 / 低增长代理）。"
+  },
+  "academic_corr_rewire": {
+   "en": "Causal rolling rendition of the correlation-regime skill's Mode 4 (rewiring leaderboard): per-asset row mean of |corr(event) - corr(calm)| where the event window is the trailing 20 bars and the calm baseline is the 120 bars immediately before it, sign-flipped and cross-sectionally z-scored so correlation-stable names rank highest. Framed as a structural risk / crowding lens, not a validated alpha: the author's public validation of the underlying regime machinery covers regime detection and attribution on historical replays only and makes no return-prediction claim for this cross-sectional form. Missing data: columns need >= 90% in-window coverage, else NaN; observed entries are demeaned and missing entries contribute zero to the correlation cross-products.",
+   "zh": "correlation-regime skill 中 Mode 4（重连排行榜）的因果滚动实现：对每个资产取 |corr(event) - corr(calm)| 的行均值，其中事件窗口为最近 20 根 bar，平静基线为紧邻其前的 120 根 bar；结果取反号后做截面 z 分数，使相关性稳定的标的排名最高。定位为结构性风险 / 拥挤度视角，而非经过验证的 alpha：作者对底层 regime 机制的公开验证仅覆盖历史回放上的 regime 识别与归因，并未对这一截面形式作出任何收益预测主张。缺失数据：各列在窗口内的覆盖率须 >= 90%，否则为 NaN；观测值先去均值，缺失值在相关系数交叉乘积中的贡献为零。"
+  },
+  "academic_high52w": {
+   "en": "George & Hwang (2004) 52-week-high momentum. Ratio of current close to the trailing 252-day maximum close, cross-sectional z-score per date for long-short ranking. Top z-scores = names trading nearest their 52-week high. Canonical 252d window; declared decay_horizon=60 due to registry schema cap (le=60); real signal horizon spans months.",
+   "zh": "George & Hwang (2004) 52 周高点动量。当前收盘价与过去 252 日最高收盘价之比，按日期做截面 z 分数，用于多空排名。z 分数最高 = 价格最接近其 52 周高点的标的。采用标准的 252d 窗口；受注册表 schema 上限（le=60）所限，声明 decay_horizon=60；实际信号周期跨越数月。"
+  },
+  "academic_hml": {
+   "en": "[PRICE PROXY] for the Fama-French (1993) HML (High Minus Low) value factor. The original definition uses book-to-market ratio from fundamental data; here we use the negative 252-day total return as a long-term reversal proxy, then cross-sectional z-score per date for long-short ranking. Top z-scores = long-term underperformers (deeper value). Canonical 252d window; declared decay_horizon=60 due to registry schema cap (le=60); real signal horizon=252.",
+   "zh": "[价格代理] 对应 Fama-French (1993) HML（High Minus Low，高减低）价值因子。原始定义使用基本面数据中的账面市值比；此处改用 252 日总收益率的负值作为长期反转代理，再按日期做截面 z 分数，用于多空排名。z 分数最高 = 长期表现落后者（价值更深）。采用标准的 252d 窗口；受注册表 schema 上限（le=60）所限，声明 decay_horizon=60；实际信号周期=252。"
+  },
+  "academic_illiq": {
+   "en": "Amihud (2002) ILLIQ. Daily |simple return| divided by dollar volume (close * volume), averaged over a trailing 21-day window, then cross-sectional z-score per date for long-short ranking. Top z-scores = least liquid names, which carry the illiquidity premium. Uses a 1-day return (one extra warmup bar over the 21-day average window).",
+   "zh": "Amihud (2002) ILLIQ。每日 |简单收益率| 除以成交额（close * volume），在过去 21 日窗口内取均值，再按日期做截面 z 分数，用于多空排名。z 分数最高 = 流动性最差的标的，承载非流动性溢价。使用 1 日收益率（比 21 日均值窗口多需一根预热 bar）。"
+  },
+  "academic_mkt_rf": {
+   "en": "[PRICE PROXY] for the Sharpe (1964) / Fama-French market factor (MKT-RF). The original definition uses value-weighted market excess returns; here we use a 21-day per-stock total return and cross-sectional z-score per date for long-short ranking. Top z-scores = strong recent winners; bottom = losers.",
+   "zh": "[价格代理] 对应 Sharpe (1964) / Fama-French 市场因子（MKT-RF）。原始定义使用市值加权的市场超额收益率；此处改用个股 21 日总收益率，并按日期做截面 z 分数，用于多空排名。z 分数最高 = 近期强势赢家；最低 = 输家。"
+  },
+  "academic_retskew": {
+   "en": "Harvey & Siddique (2000) skewness premium. Negative of the trailing 60-day sample skewness of daily simple returns, cross-sectional z-score per date for long-short ranking. Top z-scores = most negatively-skewed names, which command a return premium over lottery-like positively-skewed names. Uses a 1-day return (one extra warmup bar over the 60-day window).",
+   "zh": "Harvey & Siddique (2000) 偏度溢价。取日简单收益率过去 60 日样本偏度的负值，按日期做截面 z 分数，用于多空排名。z 分数最高 = 负偏最强的标的，相对于类彩票的正偏标的可获得收益率溢价。使用 1 日收益率（比 60 日窗口多需一根预热 bar）。"
+  },
+  "academic_rmw": {
+   "en": "[PRICE PROXY] for the Fama-French (2015) RMW (Robust Minus Weak) profitability factor. The original definition uses operating profitability from fundamental data; here we use the negative of 60-day return volatility as a low-vol-quality proxy, then cross-sectional z-score per date for long-short ranking. Top z-scores = lower vol (quality / robust).",
+   "zh": "[价格代理] 对应 Fama-French (2015) RMW（Robust Minus Weak，稳健减疲弱）盈利能力因子。原始定义使用基本面数据中的经营盈利能力；此处改用 60 日收益率波动率的负值作为低波动-质量代理，再按日期做截面 z 分数，用于多空排名。z 分数最高 = 波动率较低（质量 / 稳健）。"
+  },
+  "academic_smb": {
+   "en": "[PRICE PROXY] for the Fama-French (1993) SMB (Small Minus Big) size factor. The original definition uses market capitalization from book equity data; here we use the negative log of 60-day average dollar volume (close * volume) as a liquidity-weighted size proxy, then cross-sectional z-score per date for long-short ranking. Top z-scores = smaller / less liquid names.",
+   "zh": "[价格代理] 对应 Fama-French (1993) SMB（Small Minus Big，小减大）规模因子。原始定义使用账面权益数据中的市值；此处改用 60 日平均成交额（close * volume）对数的负值作为流动性加权的规模代理，再按日期做截面 z 分数，用于多空排名。z 分数最高 = 规模更小 / 流动性更差的标的。"
+  },
+  "academic_strev": {
+   "en": "Jegadeesh (1990) one-month reversal. Negative trailing 21-day return, cross-sectional z-score per date for long-short ranking. Top z-scores = recent one-month losers, which tend to rebound. Constructed directly from prices, matching the original definition modulo the z-score wrapper.",
+   "zh": "Jegadeesh (1990) 一个月反转。取过去 21 日收益率的负值，按日期做截面 z 分数，用于多空排名。z 分数最高 = 近一个月的输家，往往会反弹。直接由价格构造，除外层的 z 分数变换外与原始定义一致。"
+  },
+  "alpha101_019": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_024": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_032": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_036": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_037": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_039": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_048": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_052": {
+   "en": "Very long lookback (>= ~100 bars); produces NaN warmup on short panels which may trigger the >95% NaN registry guard.",
+   "zh": "回看窗口极长（>= ~100 根 bar）；在较短的面板上会产生 NaN 预热期，可能触发注册表的 >95% NaN 守卫。"
+  },
+  "alpha101_056": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization. Paper formula uses market 'cap' which is not part of the standard OHLCV panel; substituted by a constant 1.0 DataFrame. Result remains a valid factor but loses the cap-weighting term.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。论文公式使用市值 'cap'，它不属于标准 OHLCV 面板；此处以常数 1.0 的 DataFrame 替代。结果仍是有效因子，但失去了市值加权项。"
+  },
+  "alpha101_058": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_059": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_063": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_067": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_069": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_070": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_076": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_079": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_080": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_082": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_084": {
+   "en": "SignedPower with a delta(close,5) exponent can produce non-finite values when the exponent is large; non-finite outputs are clipped to NaN to satisfy the registry's no-inf invariant.",
+   "zh": "指数为 delta(close,5) 的 SignedPower 在指数很大时可能产生非有限值；非有限输出被裁剪为 NaN，以满足注册表的 no-inf（不含 inf）不变式。"
+  },
+  "alpha101_087": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_089": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_090": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_091": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_093": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_097": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "alpha101_100": {
+   "en": "Industry neutralization implemented via per-row sector group demean (panel['sector'] required). When sector tag is absent the registry rejects via SkipAlpha; the compute() also has a degraded global demean fallback. This is a partial approximation of the paper's IndClass.industry/subindustry/sector neutralization.",
+   "zh": "行业中性化通过逐行按 sector 分组去均值实现（需要 panel['sector']）。缺少 sector 标签时，注册表通过 SkipAlpha 予以拒绝；compute() 中也带有降级的全局去均值回退。这是对论文 IndClass.industry/subindustry/sector 中性化的部分近似。"
+  },
+  "fund_asset_growth": {
+   "en": "Inverse PIT-safe asset-growth factor. High asset growth is treated as aggressive investment and receives lower scores; low/negative growth receives higher scores.",
+   "zh": "反向的时点正确（PIT）资产增长因子。高资产增长被视为激进投资，得分较低；低 / 负增长得分较高。"
+  },
+  "fund_earnings_yield": {
+   "en": "Hybrid value factor using PIT-safe net income and shares diluted aligned to the daily close panel. Division by zero/non-finite market cap becomes NaN before cross-sectional scoring.",
+   "zh": "混合型价值因子：使用时点正确（PIT）的净利润与稀释后股数，并对齐到日度收盘价面板。截面打分之前，除以零 / 非有限市值的结果置为 NaN。"
+  },
+  "fund_gross_profitability": {
+   "en": "PIT-safe Novy-Marx style gross profitability from the fundamental panel; cross-sectional z-score per date with missing issuers excluded.",
+   "zh": "基于基本面面板的时点正确（PIT）Novy-Marx 式毛利盈利能力；按日期做截面 z 分数，缺失的发行人被排除在外。"
+  },
+  "fund_roe": {
+   "en": "PIT-safe return on equity from the fundamental panel; cross-sectional z-score per date. Missing statements remain NaN and are excluded from that date's cross-section.",
+   "zh": "基于基本面面板的时点正确（PIT）净资产收益率（ROE）；按日期做截面 z 分数。缺失的财务报表保持 NaN，并被排除在当日截面之外。"
+  },
+  "gtja191_001": {
+   "en": "Standard GTJA #1: lag-1 log-volume change rank vs intraday return rank, 6d corr.",
+   "zh": "标准 GTJA #1：滞后 1 期对数成交量变化的排名与日内收益率排名之间的 6 日相关系数。"
+  },
+  "gtja191_002": {
+   "en": "Daily change in close-position-within-range.",
+   "zh": "收盘价在当日高低区间内相对位置的日度变化。"
+  },
+  "gtja191_003": {
+   "en": "Wilder-style accumulation of signed daily moves over 6 days.",
+   "zh": "Wilder 式的 6 日带符号日度价格变动累加。"
+  },
+  "gtja191_004": {
+   "en": "Breakout signal: short-MA vs long-MA +/- 1 std, volume-relative tiebreaker. Output in {-1, +1}.",
+   "zh": "突破信号：短期 MA 对比长期 MA ± 1 倍标准差，以相对成交量打破平局。输出取值于 {-1, +1}。"
+  },
+  "gtja191_005": {
+   "en": "Max over 3 days of 5d corr of TSRANK(volume,5) and TSRANK(high,5).",
+   "zh": "TSRANK(volume,5) 与 TSRANK(high,5) 的 5 日相关系数在 3 日内的最大值。"
+  },
+  "gtja191_006": {
+   "en": "Sign of 4d change of weighted price; cross-sectionally ranked, negated.",
+   "zh": "加权价格 4 日变化的符号；经截面排名后取负。"
+  },
+  "gtja191_007": {
+   "en": "VWAP via A-share amount/volume (equity_cn convention).",
+   "zh": "成交量加权均价（VWAP）按 A 股成交额/成交量计算（equity_cn 惯例）。"
+  },
+  "gtja191_008": {
+   "en": "Negated rank of 4d change in mid-vwap composite.",
+   "zh": "中间价–vwap 组合价 4 日变化的排名，取负。"
+  },
+  "gtja191_009": {
+   "en": "SMA(n=7, m=2) of midpoint change times range / volume.",
+   "zh": "中间价变化乘以高低价差再除以成交量，取 SMA(n=7, m=2)。"
+  },
+  "gtja191_010": {
+   "en": "Per-day return = pct_change(1) via (close - delay(close,1))/delay(close,1).",
+   "zh": "日收益率 = pct_change(1)，按 (close - delay(close,1))/delay(close,1) 计算。"
+  },
+  "gtja191_011": {
+   "en": "Accumulated money-flow-multiplier × volume over 6 days.",
+   "zh": "资金流乘数 × 成交量在 6 日内的累加。"
+  },
+  "gtja191_012": {
+   "en": "Open-minus-10d-vwap rank times negative rank of |close-vwap|.",
+   "zh": "开盘价减 10 日 vwap 均值之差的排名，乘以 |close-vwap| 的负排名。"
+  },
+  "gtja191_013": {
+   "en": "Geometric mean of high/low minus vwap.",
+   "zh": "最高价与最低价的几何平均减去 vwap。"
+  },
+  "gtja191_014": {
+   "en": "Simple 5d momentum = delta(close, 5).",
+   "zh": "简单 5 日动量 = delta(close, 5)。"
+  },
+  "gtja191_015": {
+   "en": "Overnight gap return.",
+   "zh": "隔夜跳空收益率。"
+  },
+  "gtja191_016": {
+   "en": "Max over 5d of rank of rolling rank-volume vs rank-vwap correlation.",
+   "zh": "成交量排名与 vwap 排名的滚动相关系数，其排名在 5 日内的最大值。"
+  },
+  "gtja191_017": {
+   "en": "rank(vwap - 15d max(vwap)) is non-positive; we use signed_power for safety.",
+   "zh": "rank(vwap - 15 日 max(vwap)) 非正；为稳妥起见使用 signed_power（带符号幂）。"
+  },
+  "gtja191_018": {
+   "en": "5d price ratio.",
+   "zh": "5 日价格比值。"
+  },
+  "gtja191_019": {
+   "en": "Piecewise 5d momentum normalized differently in up/down regimes.",
+   "zh": "分段 5 日动量：上涨与下跌区间采用不同的归一化方式。"
+  },
+  "gtja191_020": {
+   "en": "6d return in pct.",
+   "zh": "6 日收益率（百分比）。"
+  },
+  "gtja191_021": {
+   "en": "Rolling 6-day slope of MA6(close) vs time index. REGBETA proxied by ts_cov / ts_std**2.",
+   "zh": "MA6(close) 对时间索引的滚动 6 日斜率。REGBETA 以 ts_cov / ts_std**2 近似代替。"
+  },
+  "gtja191_022": {
+   "en": "SMA(12, m=1) of 3-day-difference in price-deviation-from-MA6.",
+   "zh": "价格相对 MA6 偏离度的 3 日差分，取 SMA(12, m=1)。"
+  },
+  "gtja191_023": {
+   "en": "Up-volatility share. SMA(20, m=1) of STD(20) over up/down days.",
+   "zh": "上涨波动率占比。对上涨/下跌日的 STD(20) 取 SMA(20, m=1)。"
+  },
+  "gtja191_024": {
+   "en": "SMA(5, m=1) of 5d delta(close).",
+   "zh": "5 日 delta(close) 的 SMA(5, m=1)。"
+  },
+  "gtja191_025": {
+   "en": "Long-window RET sum approximated with 60d cap (warmup feasibility); see notes.",
+   "zh": "长窗口 RET 求和以 60 日上限近似（为保证预热期可行）；详见说明。"
+  },
+  "gtja191_026": {
+   "en": "230d corr approximated with 30d window; see notes.",
+   "zh": "230 日相关系数以 30 日窗口近似；详见说明。"
+  },
+  "gtja191_027": {
+   "en": "WMA proxied by decay_linear.",
+   "zh": "WMA 以 decay_linear 近似代替。"
+  },
+  "gtja191_028": {
+   "en": "Stochastic-like indicator double-smoothed.",
+   "zh": "经双重平滑的类随机（Stochastic）指标。"
+  },
+  "gtja191_029": {
+   "en": "6d return times current volume.",
+   "zh": "6 日收益率乘以当日成交量。"
+  },
+  "gtja191_030": {
+   "en": "Multi-factor REGRESI not implementable in pure-fn zoo; degraded to WMA of squared daily return (idio proxy). See notes.",
+   "zh": "多因子 REGRESI 无法在纯函数 zoo 中实现；退化为日收益率平方的 WMA（特质项代理）。详见说明。"
+  },
+  "gtja191_031": {
+   "en": "Bias-12: deviation of close from MA12 in pct.",
+   "zh": "Bias-12（乖离率）：收盘价相对 MA12 的偏离百分比。"
+  },
+  "gtja191_032": {
+   "en": "Negated 3d sum of rank-corr(rank(high), rank(volume), 3).",
+   "zh": "rank-corr(rank(high), rank(volume), 3) 的 3 日求和，取负。"
+  },
+  "gtja191_033": {
+   "en": "240d/20d long-window approximated with 60d/20d (warmup feasibility).",
+   "zh": "240 日/20 日长窗口以 60 日/20 日近似（为保证预热期可行）。"
+  },
+  "gtja191_034": {
+   "en": "MA12 over close.",
+   "zh": "MA12 除以收盘价。"
+  },
+  "gtja191_035": {
+   "en": "Element-wise min of two ranks, negated.",
+   "zh": "两个排名的逐元素最小值，取负。"
+  },
+  "gtja191_036": {
+   "en": "Rolling 6d corr summed over 2 days then ranked.",
+   "zh": "滚动 6 日相关系数在 2 日内求和后再排名。"
+  },
+  "gtja191_037": {
+   "en": "Change over 10d of product (sum(open,5)*sum(ret,5)).",
+   "zh": "乘积 (sum(open,5)*sum(ret,5)) 的 10 日变化。"
+  },
+  "gtja191_038": {
+   "en": "When current high > MA20(high), output -delta(high,2); else 0.",
+   "zh": "当日最高价 > MA20(high) 时输出 -delta(high,2)；否则为 0。"
+  },
+  "gtja191_039": {
+   "en": "180d / 37d windows approximated with 30d / 10d. See notes.",
+   "zh": "180 日 / 37 日窗口以 30 日 / 10 日近似。详见说明。"
+  },
+  "gtja191_040": {
+   "en": "Up-volume vs down-volume ratio over 26 days.",
+   "zh": "26 日内上涨日成交量与下跌日成交量之比。"
+  },
+  "gtja191_041": {
+   "en": "5d max of 3d delta(vwap), ranked, negated.",
+   "zh": "3 日 delta(vwap) 的 5 日最大值，排名后取负。"
+  },
+  "gtja191_042": {
+   "en": "Negative rank of 10d std(high) times 10d corr(high, volume).",
+   "zh": "10 日 std(high) 的负排名乘以 10 日 corr(high, volume)。"
+  },
+  "gtja191_043": {
+   "en": "Signed volume accumulation over 6 days.",
+   "zh": "6 日内带符号成交量的累加。"
+  },
+  "gtja191_044": {
+   "en": "Sum of two TSRANK terms.",
+   "zh": "两个 TSRANK 项之和。"
+  },
+  "gtja191_045": {
+   "en": "150d MA volume approximated with 30d window.",
+   "zh": "150 日成交量 MA 以 30 日窗口近似。"
+  },
+  "gtja191_046": {
+   "en": "Mean of four MA windows over price.",
+   "zh": "四个窗口 MA 的平均值除以价格。"
+  },
+  "gtja191_047": {
+   "en": "Williams %R style indicator smoothed with SMA(9,1).",
+   "zh": "Williams %R 类指标，以 SMA(9,1) 平滑。"
+  },
+  "gtja191_048": {
+   "en": "Rank of 3d signed momentum sum times 5d/20d volume ratio, negated.",
+   "zh": "3 日带符号动量之和的排名乘以 5 日/20 日成交量比，取负。"
+  },
+  "gtja191_049": {
+   "en": "Down-side range as share of total range over 12 days.",
+   "zh": "12 日内下行波幅占总波幅的比例。"
+  },
+  "gtja191_050": {
+   "en": "Signed version of #49.",
+   "zh": "#49 的带符号版本。"
+  },
+  "gtja191_051": {
+   "en": "Up-range share over 12 days.",
+   "zh": "12 日内上行波幅占比。"
+  },
+  "gtja191_052": {
+   "en": "Bull power / bear power ratio over 26 days.",
+   "zh": "26 日内多头力量 / 空头力量之比。"
+  },
+  "gtja191_053": {
+   "en": "Pct of up-days in 12d window.",
+   "zh": "12 日窗口内上涨日的百分比。"
+  },
+  "gtja191_054": {
+   "en": "Negated rank of (std|c-o|,10) + (c-o) + corr(c,o,10).",
+   "zh": "(std|c-o|,10) + (c-o) + corr(c,o,10) 的排名，取负。"
+  },
+  "gtja191_055": {
+   "en": "Sumof complex per-day score over 20 days; numerator simplified.",
+   "zh": "逐日复合得分的 20 日求和；分子已简化。"
+  },
+  "gtja191_056": {
+   "en": "Boolean comparator returns 1/0; 40d mean truncated.",
+   "zh": "布尔比较返回 1/0；40 日均值窗口已截短。"
+  },
+  "gtja191_057": {
+   "en": "KDJ %K-style indicator with SMA(3,1) smoothing.",
+   "zh": "KDJ %K 类指标，以 SMA(3,1) 平滑。"
+  },
+  "gtja191_058": {
+   "en": "Pct of up-days in 20d window.",
+   "zh": "20 日窗口内上涨日的百分比。"
+  },
+  "gtja191_059": {
+   "en": "Like alpha #3 but with 20d window.",
+   "zh": "与 alpha #3 相同，但窗口为 20 日。"
+  },
+  "gtja191_060": {
+   "en": "20-day version of alpha #11.",
+   "zh": "alpha #11 的 20 日版本。"
+  },
+  "gtja191_061": {
+   "en": "80d mean volume approximated with 30d.",
+   "zh": "80 日平均成交量以 30 日近似。"
+  },
+  "gtja191_062": {
+   "en": "Negated 5d corr(high, rank(volume)).",
+   "zh": "5 日 corr(high, rank(volume))，取负。"
+  },
+  "gtja191_063": {
+   "en": "RSI-6 style.",
+   "zh": "RSI-6 类指标。"
+  },
+  "gtja191_064": {
+   "en": "60d mean truncated to 10d; ts_max window 13→4 and decay_linear 14→6 for warmup feasibility.",
+   "zh": "60 日均值截短为 10 日；ts_max 窗口 13→4、decay_linear 14→6，以保证预热期可行。"
+  },
+  "gtja191_065": {
+   "en": "MA6 over close.",
+   "zh": "MA6 除以收盘价。"
+  },
+  "gtja191_066": {
+   "en": "Bias-6 pct.",
+   "zh": "Bias-6（乖离率），百分比。"
+  },
+  "gtja191_067": {
+   "en": "RSI-24 style.",
+   "zh": "RSI-24 类指标。"
+  },
+  "gtja191_068": {
+   "en": "Like alpha #9 but with SMA(15, m=2).",
+   "zh": "与 alpha #9 相同，但使用 SMA(15, m=2)。"
+  },
+  "gtja191_069": {
+   "en": "DTM/DBM as in GTJA spec.",
+   "zh": "DTM/DBM 按 GTJA 规范定义。"
+  },
+  "gtja191_070": {
+   "en": "6d std of amount (turnover).",
+   "zh": "amount（成交额，turnover）的 6 日标准差。"
+  },
+  "gtja191_071": {
+   "en": "Bias-24.",
+   "zh": "Bias-24（乖离率）。"
+  },
+  "gtja191_072": {
+   "en": "Williams %R style with SMA(15,1).",
+   "zh": "Williams %R 类指标，使用 SMA(15,1)。"
+  },
+  "gtja191_073": {
+   "en": "Composite of decay-linear of close-volume corr minus another decay term.",
+   "zh": "收盘价–成交量相关系数的线性衰减加权项减去另一线性衰减加权项的组合。"
+  },
+  "gtja191_074": {
+   "en": "40d MA volume truncated to 10d; SUM windows kept at 20.",
+   "zh": "40 日成交量 MA 截短为 10 日；SUM 窗口保持为 20。"
+  },
+  "gtja191_075": {
+   "en": "Benchmark unavailable in zoo panel; degraded to row-mean(close) as benchmark proxy. Window 50→20. See notes.",
+   "zh": "zoo 面板中无基准数据；退化为以 row-mean(close) 作为基准代理。窗口 50→20。详见说明。"
+  },
+  "gtja191_076": {
+   "en": "Coefficient-of-variation of |daily return|/volume over 20 days.",
+   "zh": "|日收益率|/成交量 在 20 日内的变异系数。"
+  },
+  "gtja191_077": {
+   "en": "40d MA truncated to 30d.",
+   "zh": "40 日 MA 截短为 30 日。"
+  },
+  "gtja191_078": {
+   "en": "CCI-12 style.",
+   "zh": "CCI-12 类指标。"
+  },
+  "gtja191_079": {
+   "en": "RSI-12.",
+   "zh": "RSI-12 指标。"
+  },
+  "gtja191_080": {
+   "en": "5d volume change pct.",
+   "zh": "成交量 5 日变化百分比。"
+  },
+  "gtja191_081": {
+   "en": "SMA(21, m=2) of volume.",
+   "zh": "成交量的 SMA(21, m=2)。"
+  },
+  "gtja191_082": {
+   "en": "Williams %R with SMA(20,1).",
+   "zh": "Williams %R，使用 SMA(20,1)。"
+  },
+  "gtja191_083": {
+   "en": "Negated rank of 5d cov(rank(high), rank(volume)).",
+   "zh": "5 日 cov(rank(high), rank(volume)) 的排名，取负。"
+  },
+  "gtja191_084": {
+   "en": "20d version of alpha #43.",
+   "zh": "alpha #43 的 20 日版本。"
+  },
+  "gtja191_085": {
+   "en": "Vol-adjusted contrarian momentum.",
+   "zh": "成交量调整的逆向动量。"
+  },
+  "gtja191_086": {
+   "en": "Returns -1 / +1 / -delta depending on second-derivative thresholds.",
+   "zh": "依据二阶导数阈值返回 -1 / +1 / -delta。"
+  },
+  "gtja191_087": {
+   "en": "Sum of two decay-linear terms, negated.",
+   "zh": "两个线性衰减加权项之和，取负。"
+  },
+  "gtja191_088": {
+   "en": "20d return pct.",
+   "zh": "20 日收益率（百分比）。"
+  },
+  "gtja191_089": {
+   "en": "MACD-like signal.",
+   "zh": "类 MACD 信号。"
+  },
+  "gtja191_090": {
+   "en": "Negated rank of 5d corr(rank vwap, rank volume).",
+   "zh": "5 日 corr(rank vwap, rank volume) 的排名，取负。"
+  },
+  "gtja191_091": {
+   "en": "40d mean truncated to 30d.",
+   "zh": "40 日均值截短为 30 日。"
+  },
+  "gtja191_092": {
+   "en": "180d mean truncated to 30d.",
+   "zh": "180 日均值截短为 30 日。"
+  },
+  "gtja191_093": {
+   "en": "Sum of downside range moves over 20 days.",
+   "zh": "20 日内下行波幅之和。"
+  },
+  "gtja191_094": {
+   "en": "30d signed volume.",
+   "zh": "30 日带符号成交量。"
+  },
+  "gtja191_095": {
+   "en": "20d std of amount.",
+   "zh": "成交额（amount）的 20 日标准差。"
+  },
+  "gtja191_096": {
+   "en": "KDJ %D-style double smoothed.",
+   "zh": "KDJ %D 类指标，经双重平滑。"
+  },
+  "gtja191_097": {
+   "en": "10d std of volume.",
+   "zh": "成交量的 10 日标准差。"
+  },
+  "gtja191_098": {
+   "en": "100d windows truncated to 30d for warmup feasibility.",
+   "zh": "为保证预热期可行，100 日窗口截短为 30 日。"
+  },
+  "gtja191_099": {
+   "en": "Negated rank of 5d cov(rank close, rank volume).",
+   "zh": "5 日 cov(rank close, rank volume) 的排名，取负。"
+  },
+  "gtja191_100": {
+   "en": "20d std of volume.",
+   "zh": "成交量的 20 日标准差。"
+  },
+  "gtja191_101": {
+   "en": "Inequality cast to float via .astype('float64').",
+   "zh": "不等式结果通过 .astype('float64') 转为 float。"
+  },
+  "gtja191_102": {
+   "en": "SMA(x,n,m) -> x.ewm(alpha=m/n, adjust=False).mean().",
+   "zh": "SMA(x,n,m) 映射为 x.ewm(alpha=m/n, adjust=False).mean()。"
+  },
+  "gtja191_103": {
+   "en": "Days since the most recent extreme; today=0. Ties choose the latest bar (port policy).",
+   "zh": "距最近一次极值的天数；当日记为 0。并列时取最近的一根 bar（移植实现的约定）。"
+  },
+  "gtja191_108": {
+   "en": "x^y interpreted as x ** y after rank; both terms in (0,1].",
+   "zh": "x^y 在排名之后按 x ** y 解释；两项均位于 (0,1]。"
+  },
+  "gtja191_109": {
+   "en": "SMA -> ewm(alpha=2/10).",
+   "zh": "SMA 映射为 ewm(alpha=2/10)。"
+  },
+  "gtja191_115": {
+   "en": "x^y -> np.power after rank.",
+   "zh": "x^y 在排名之后映射为 np.power。"
+  },
+  "gtja191_116": {
+   "en": "Rolling OLS slope vs. linear index; cov(c,t,20)/var(t,20).",
+   "zh": "对线性索引的滚动 OLS 斜率；cov(c,t,20)/var(t,20)。"
+  },
+  "gtja191_117": {
+   "en": "ret = close/delay(close,1) - 1.",
+   "zh": "收益率 ret = close/delay(close,1) - 1。"
+  },
+  "gtja191_122": {
+   "en": "Triple SMA of log close.",
+   "zh": "对数收盘价的三重 SMA。"
+  },
+  "gtja191_131": {
+   "en": "DELAT in report typo = DELTA.",
+   "zh": "研报中的 DELAT 为笔误，即 DELTA。"
+  },
+  "gtja191_133": {
+   "en": "Days since the most recent extreme; today=0. Ties choose the latest bar (port policy).",
+   "zh": "距最近一次极值的天数；当日记为 0。并列时取最近的一根 bar（移植实现的约定）。"
+  },
+  "gtja191_136": {
+   "en": "ret = close/delay(close,1) - 1.",
+   "zh": "收益率 ret = close/delay(close,1) - 1。"
+  },
+  "gtja191_137": {
+   "en": "Transcribed from the standard 137 implementation; piecewise denominator.",
+   "zh": "转录自 137 的标准实现；分母为分段形式。"
+  },
+  "gtja191_140": {
+   "en": "MIN(a,b) elementwise -> np.minimum.",
+   "zh": "逐元素 MIN(a,b) 映射为 np.minimum。"
+  },
+  "gtja191_143": {
+   "en": "Recursive SELF unrolled to cumulative product of (1 + up_return) since series start.",
+   "zh": "递归的 SELF 展开为自序列起点以来 (1 + up_return) 的累积乘积。"
+  },
+  "gtja191_144": {
+   "en": "SUMIF -> (x*cond).rolling(n).sum(); COUNT -> cond.rolling(n).sum().",
+   "zh": "SUMIF 映射为 (x*cond).rolling(n).sum()；COUNT 映射为 cond.rolling(n).sum()。"
+  },
+  "gtja191_146": {
+   "en": "Standardised return deviation; SMA(.,61,2)=ewm(alpha=2/61).",
+   "zh": "标准化的收益率偏离；SMA(.,61,2)=ewm(alpha=2/61)。"
+  },
+  "gtja191_147": {
+   "en": "Rolling OLS slope of MA12 against linear index, window 12.",
+   "zh": "MA12 对线性索引的滚动 OLS 斜率，窗口 12。"
+  },
+  "gtja191_149": {
+   "en": "Downside beta vs. benchmark; uses fallback cross-sectional mean if benchmark_close missing.",
+   "zh": "相对基准的下行贝塔；若缺少 benchmark_close，则回退为截面均值。"
+  },
+  "gtja191_154": {
+   "en": "Original returns boolean; we cast to float and multiply by -1.",
+   "zh": "原公式返回布尔值；此处转为 float 并乘以 -1。"
+  },
+  "gtja191_156": {
+   "en": "MAX elementwise -> np.maximum.",
+   "zh": "逐元素 MAX 映射为 np.maximum。"
+  },
+  "gtja191_157": {
+   "en": "PROD(.,1) is identity; we use it directly.",
+   "zh": "PROD(.,1) 为恒等运算；此处直接使用。"
+  },
+  "gtja191_161": {
+   "en": "Average True Range (12).",
+   "zh": "平均真实波幅（12）。"
+  },
+  "gtja191_162": {
+   "en": "RSI-style normalised.",
+   "zh": "RSI 类指标，经归一化。"
+  },
+  "gtja191_165": {
+   "en": "SUMAC = expanding cumulative sum approximated by rolling 48-day cumulative sum.",
+   "zh": "SUMAC = 扩展窗口累计和，以滚动 48 日累计和近似。"
+  },
+  "gtja191_166": {
+   "en": "Skewness-style; constants from report.",
+   "zh": "偏度类指标；常数取自研报。"
+  },
+  "gtja191_172": {
+   "en": "Wilder's ADX-style indicator; mean over last 6 bars.",
+   "zh": "Wilder 的 ADX 类指标；取最近 6 根 bar 的均值。"
+  },
+  "gtja191_175": {
+   "en": "ATR(6).",
+   "zh": "ATR(6)（平均真实波幅）。"
+  },
+  "gtja191_177": {
+   "en": "Days since the most recent extreme; today=0. Ties choose the latest bar (port policy).",
+   "zh": "距最近一次极值的天数；当日记为 0。并列时取最近的一根 bar（移植实现的约定）。"
+  },
+  "gtja191_181": {
+   "en": "Benchmark falls back to cross-sectional mean of close.",
+   "zh": "基准回退为收盘价的截面均值。"
+  },
+  "gtja191_182": {
+   "en": "Benchmark falls back to cross-sectional mean of close.",
+   "zh": "基准回退为收盘价的截面均值。"
+  },
+  "gtja191_186": {
+   "en": "alpha172 averaged with its 6-day lag.",
+   "zh": "alpha172 与其滞后 6 日的值取平均。"
+  },
+  "gtja191_190": {
+   "en": "Complex log of ratio of conditional squared deviations.",
+   "zh": "条件平方偏差之比的对数（表达式较复杂）。"
+  },
+  "qlib158_cntd10": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~30% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~30% 起）。"
+  },
+  "qlib158_cntd20": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~15% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~15% 起）。"
+  },
+  "qlib158_cntd30": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~10% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~10% 起）。"
+  },
+  "qlib158_cntd5": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~40% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~40% 起）。"
+  },
+  "qlib158_cntd60": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~5% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~5% 起）。"
+  },
+  "qlib158_cntn10": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~30% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~30% 起）。"
+  },
+  "qlib158_cntn20": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~15% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~15% 起）。"
+  },
+  "qlib158_cntn30": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~10% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~10% 起）。"
+  },
+  "qlib158_cntn5": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~40% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~40% 起）。"
+  },
+  "qlib158_cntn60": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~5% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~5% 起）。"
+  },
+  "qlib158_cntp10": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~30% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~30% 起）。"
+  },
+  "qlib158_cntp20": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~15% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~15% 起）。"
+  },
+  "qlib158_cntp30": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~10% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~10% 起）。"
+  },
+  "qlib158_cntp5": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~40% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~40% 起）。"
+  },
+  "qlib158_cntp60": {
+   "en": "A missing close leaves window+1 rolling rows NaN; on sparse panels this can trigger the >95% NaN registry guard (from ~5% missing bars).",
+   "zh": "一个缺失的收盘价会使 window+1 行滚动结果为 NaN；在稀疏面板上，这可能触发注册表的 >95% NaN 守卫（缺失 bar 达 ~5% 起）。"
+  }
+ },
+ "nicknames": {
+  "academic_bab": {
+   "en": "Frazzini-Pedersen 2014 betting-against-beta",
+   "zh": "Frazzini-Pedersen 2014 低贝塔策略（BAB）"
+  },
+  "academic_carhart_mom": {
+   "en": "Carhart 1997 momentum — 12m-1m return",
+   "zh": "Carhart 1997 动量 — 12m-1m 收益率"
+  },
+  "academic_cma": {
+   "en": "[PRICE PROXY] FF2015 CMA — investment via inverse volume growth",
+   "zh": "[价格代理] FF2015 CMA — 以反向成交量增长代理投资"
+  },
+  "academic_corr_rewire": {
+   "en": "correlation-rewiring stability score",
+   "zh": "相关性重连稳定性得分"
+  },
+  "academic_high52w": {
+   "en": "George-Hwang 2004 52-week-high momentum",
+   "zh": "George-Hwang 2004 52 周高点动量"
+  },
+  "academic_hml": {
+   "en": "[PRICE PROXY] FF1993 HML — value via inverse 252d return",
+   "zh": "[价格代理] FF1993 HML — 以反向 252d 收益率代理价值"
+  },
+  "academic_illiq": {
+   "en": "Amihud 2002 illiquidity — |return| per dollar volume",
+   "zh": "Amihud 2002 非流动性 — 单位成交额的 |收益率|"
+  },
+  "academic_mkt_rf": {
+   "en": "[PRICE PROXY] Market factor (Sharpe 1964) — 21d demeaned return",
+   "zh": "[价格代理] 市场因子（Sharpe 1964）— 21d 去均值收益率"
+  },
+  "academic_retskew": {
+   "en": "Harvey-Siddique 2000 return skewness — inverse 60d skew",
+   "zh": "Harvey-Siddique 2000 收益率偏度 — 反向 60d 偏度"
+  },
+  "academic_rmw": {
+   "en": "[PRICE PROXY] FF2015 RMW — quality via inverse 60d volatility",
+   "zh": "[价格代理] FF2015 RMW — 以反向 60d 波动率代理质量"
+  },
+  "academic_smb": {
+   "en": "[PRICE PROXY] FF1993 SMB — small-minus-big via inverse dollar-volume",
+   "zh": "[价格代理] FF1993 SMB — 以反向成交额代理小减大"
+  },
+  "academic_strev": {
+   "en": "Jegadeesh 1990 short-term reversal — inverse 21d return",
+   "zh": "Jegadeesh 1990 短期反转 — 反向 21d 收益率"
+  },
+  "alpha101_001": {
+   "en": "Kakushadze Alpha #1",
+   "zh": "Kakushadze 因子 Alpha #1"
+  },
+  "alpha101_002": {
+   "en": "Kakushadze Alpha #2",
+   "zh": "Kakushadze 因子 Alpha #2"
+  },
+  "alpha101_003": {
+   "en": "Kakushadze Alpha #3",
+   "zh": "Kakushadze 因子 Alpha #3"
+  },
+  "alpha101_004": {
+   "en": "Kakushadze Alpha #4",
+   "zh": "Kakushadze 因子 Alpha #4"
+  },
+  "alpha101_005": {
+   "en": "Kakushadze Alpha #5",
+   "zh": "Kakushadze 因子 Alpha #5"
+  },
+  "alpha101_006": {
+   "en": "Kakushadze Alpha #6",
+   "zh": "Kakushadze 因子 Alpha #6"
+  },
+  "alpha101_007": {
+   "en": "Kakushadze Alpha #7",
+   "zh": "Kakushadze 因子 Alpha #7"
+  },
+  "alpha101_008": {
+   "en": "Kakushadze Alpha #8",
+   "zh": "Kakushadze 因子 Alpha #8"
+  },
+  "alpha101_009": {
+   "en": "Kakushadze Alpha #9",
+   "zh": "Kakushadze 因子 Alpha #9"
+  },
+  "alpha101_010": {
+   "en": "Kakushadze Alpha #10",
+   "zh": "Kakushadze 因子 Alpha #10"
+  },
+  "alpha101_011": {
+   "en": "Kakushadze Alpha #11",
+   "zh": "Kakushadze 因子 Alpha #11"
+  },
+  "alpha101_012": {
+   "en": "Kakushadze Alpha #12",
+   "zh": "Kakushadze 因子 Alpha #12"
+  },
+  "alpha101_013": {
+   "en": "Kakushadze Alpha #13",
+   "zh": "Kakushadze 因子 Alpha #13"
+  },
+  "alpha101_014": {
+   "en": "Kakushadze Alpha #14",
+   "zh": "Kakushadze 因子 Alpha #14"
+  },
+  "alpha101_015": {
+   "en": "Kakushadze Alpha #15",
+   "zh": "Kakushadze 因子 Alpha #15"
+  },
+  "alpha101_016": {
+   "en": "Kakushadze Alpha #16",
+   "zh": "Kakushadze 因子 Alpha #16"
+  },
+  "alpha101_017": {
+   "en": "Kakushadze Alpha #17",
+   "zh": "Kakushadze 因子 Alpha #17"
+  },
+  "alpha101_018": {
+   "en": "Kakushadze Alpha #18",
+   "zh": "Kakushadze 因子 Alpha #18"
+  },
+  "alpha101_019": {
+   "en": "Kakushadze Alpha #19",
+   "zh": "Kakushadze 因子 Alpha #19"
+  },
+  "alpha101_020": {
+   "en": "Kakushadze Alpha #20",
+   "zh": "Kakushadze 因子 Alpha #20"
+  },
+  "alpha101_021": {
+   "en": "Kakushadze Alpha #21",
+   "zh": "Kakushadze 因子 Alpha #21"
+  },
+  "alpha101_022": {
+   "en": "Kakushadze Alpha #22",
+   "zh": "Kakushadze 因子 Alpha #22"
+  },
+  "alpha101_023": {
+   "en": "Kakushadze Alpha #23",
+   "zh": "Kakushadze 因子 Alpha #23"
+  },
+  "alpha101_024": {
+   "en": "Kakushadze Alpha #24",
+   "zh": "Kakushadze 因子 Alpha #24"
+  },
+  "alpha101_025": {
+   "en": "Kakushadze Alpha #25",
+   "zh": "Kakushadze 因子 Alpha #25"
+  },
+  "alpha101_026": {
+   "en": "Kakushadze Alpha #26",
+   "zh": "Kakushadze 因子 Alpha #26"
+  },
+  "alpha101_027": {
+   "en": "Kakushadze Alpha #27",
+   "zh": "Kakushadze 因子 Alpha #27"
+  },
+  "alpha101_028": {
+   "en": "Kakushadze Alpha #28",
+   "zh": "Kakushadze 因子 Alpha #28"
+  },
+  "alpha101_029": {
+   "en": "Kakushadze Alpha #29",
+   "zh": "Kakushadze 因子 Alpha #29"
+  },
+  "alpha101_030": {
+   "en": "Kakushadze Alpha #30",
+   "zh": "Kakushadze 因子 Alpha #30"
+  },
+  "alpha101_031": {
+   "en": "Kakushadze Alpha #31",
+   "zh": "Kakushadze 因子 Alpha #31"
+  },
+  "alpha101_032": {
+   "en": "Kakushadze Alpha #32",
+   "zh": "Kakushadze 因子 Alpha #32"
+  },
+  "alpha101_033": {
+   "en": "Kakushadze Alpha #33",
+   "zh": "Kakushadze 因子 Alpha #33"
+  },
+  "alpha101_034": {
+   "en": "Kakushadze Alpha #34",
+   "zh": "Kakushadze 因子 Alpha #34"
+  },
+  "alpha101_035": {
+   "en": "Kakushadze Alpha #35",
+   "zh": "Kakushadze 因子 Alpha #35"
+  },
+  "alpha101_036": {
+   "en": "Kakushadze Alpha #36",
+   "zh": "Kakushadze 因子 Alpha #36"
+  },
+  "alpha101_037": {
+   "en": "Kakushadze Alpha #37",
+   "zh": "Kakushadze 因子 Alpha #37"
+  },
+  "alpha101_038": {
+   "en": "Kakushadze Alpha #38",
+   "zh": "Kakushadze 因子 Alpha #38"
+  },
+  "alpha101_039": {
+   "en": "Kakushadze Alpha #39",
+   "zh": "Kakushadze 因子 Alpha #39"
+  },
+  "alpha101_040": {
+   "en": "Kakushadze Alpha #40",
+   "zh": "Kakushadze 因子 Alpha #40"
+  },
+  "alpha101_041": {
+   "en": "Kakushadze Alpha #41",
+   "zh": "Kakushadze 因子 Alpha #41"
+  },
+  "alpha101_042": {
+   "en": "Kakushadze Alpha #42",
+   "zh": "Kakushadze 因子 Alpha #42"
+  },
+  "alpha101_043": {
+   "en": "Kakushadze Alpha #43",
+   "zh": "Kakushadze 因子 Alpha #43"
+  },
+  "alpha101_044": {
+   "en": "Kakushadze Alpha #44",
+   "zh": "Kakushadze 因子 Alpha #44"
+  },
+  "alpha101_045": {
+   "en": "Kakushadze Alpha #45",
+   "zh": "Kakushadze 因子 Alpha #45"
+  },
+  "alpha101_046": {
+   "en": "Kakushadze Alpha #46",
+   "zh": "Kakushadze 因子 Alpha #46"
+  },
+  "alpha101_047": {
+   "en": "Kakushadze Alpha #47",
+   "zh": "Kakushadze 因子 Alpha #47"
+  },
+  "alpha101_048": {
+   "en": "Kakushadze Alpha #48",
+   "zh": "Kakushadze 因子 Alpha #48"
+  },
+  "alpha101_049": {
+   "en": "Kakushadze Alpha #49",
+   "zh": "Kakushadze 因子 Alpha #49"
+  },
+  "alpha101_050": {
+   "en": "Kakushadze Alpha #50",
+   "zh": "Kakushadze 因子 Alpha #50"
+  },
+  "alpha101_051": {
+   "en": "Kakushadze Alpha #51",
+   "zh": "Kakushadze 因子 Alpha #51"
+  },
+  "alpha101_052": {
+   "en": "Kakushadze Alpha #52",
+   "zh": "Kakushadze 因子 Alpha #52"
+  },
+  "alpha101_053": {
+   "en": "Kakushadze Alpha #53",
+   "zh": "Kakushadze 因子 Alpha #53"
+  },
+  "alpha101_054": {
+   "en": "Kakushadze Alpha #54",
+   "zh": "Kakushadze 因子 Alpha #54"
+  },
+  "alpha101_055": {
+   "en": "Kakushadze Alpha #55",
+   "zh": "Kakushadze 因子 Alpha #55"
+  },
+  "alpha101_056": {
+   "en": "Kakushadze Alpha #56",
+   "zh": "Kakushadze 因子 Alpha #56"
+  },
+  "alpha101_057": {
+   "en": "Kakushadze Alpha #57",
+   "zh": "Kakushadze 因子 Alpha #57"
+  },
+  "alpha101_058": {
+   "en": "Kakushadze Alpha #58",
+   "zh": "Kakushadze 因子 Alpha #58"
+  },
+  "alpha101_059": {
+   "en": "Kakushadze Alpha #59",
+   "zh": "Kakushadze 因子 Alpha #59"
+  },
+  "alpha101_060": {
+   "en": "Kakushadze Alpha #60",
+   "zh": "Kakushadze 因子 Alpha #60"
+  },
+  "alpha101_061": {
+   "en": "Kakushadze Alpha #61",
+   "zh": "Kakushadze 因子 Alpha #61"
+  },
+  "alpha101_062": {
+   "en": "Kakushadze Alpha #62",
+   "zh": "Kakushadze 因子 Alpha #62"
+  },
+  "alpha101_063": {
+   "en": "Kakushadze Alpha #63",
+   "zh": "Kakushadze 因子 Alpha #63"
+  },
+  "alpha101_064": {
+   "en": "Kakushadze Alpha #64",
+   "zh": "Kakushadze 因子 Alpha #64"
+  },
+  "alpha101_065": {
+   "en": "Kakushadze Alpha #65",
+   "zh": "Kakushadze 因子 Alpha #65"
+  },
+  "alpha101_066": {
+   "en": "Kakushadze Alpha #66",
+   "zh": "Kakushadze 因子 Alpha #66"
+  },
+  "alpha101_067": {
+   "en": "Kakushadze Alpha #67",
+   "zh": "Kakushadze 因子 Alpha #67"
+  },
+  "alpha101_068": {
+   "en": "Kakushadze Alpha #68",
+   "zh": "Kakushadze 因子 Alpha #68"
+  },
+  "alpha101_069": {
+   "en": "Kakushadze Alpha #69",
+   "zh": "Kakushadze 因子 Alpha #69"
+  },
+  "alpha101_070": {
+   "en": "Kakushadze Alpha #70",
+   "zh": "Kakushadze 因子 Alpha #70"
+  },
+  "alpha101_071": {
+   "en": "Kakushadze Alpha #71",
+   "zh": "Kakushadze 因子 Alpha #71"
+  },
+  "alpha101_072": {
+   "en": "Kakushadze Alpha #72",
+   "zh": "Kakushadze 因子 Alpha #72"
+  },
+  "alpha101_073": {
+   "en": "Kakushadze Alpha #73",
+   "zh": "Kakushadze 因子 Alpha #73"
+  },
+  "alpha101_074": {
+   "en": "Kakushadze Alpha #74",
+   "zh": "Kakushadze 因子 Alpha #74"
+  },
+  "alpha101_075": {
+   "en": "Kakushadze Alpha #75",
+   "zh": "Kakushadze 因子 Alpha #75"
+  },
+  "alpha101_076": {
+   "en": "Kakushadze Alpha #76",
+   "zh": "Kakushadze 因子 Alpha #76"
+  },
+  "alpha101_077": {
+   "en": "Kakushadze Alpha #77",
+   "zh": "Kakushadze 因子 Alpha #77"
+  },
+  "alpha101_078": {
+   "en": "Kakushadze Alpha #78",
+   "zh": "Kakushadze 因子 Alpha #78"
+  },
+  "alpha101_079": {
+   "en": "Kakushadze Alpha #79",
+   "zh": "Kakushadze 因子 Alpha #79"
+  },
+  "alpha101_080": {
+   "en": "Kakushadze Alpha #80",
+   "zh": "Kakushadze 因子 Alpha #80"
+  },
+  "alpha101_081": {
+   "en": "Kakushadze Alpha #81",
+   "zh": "Kakushadze 因子 Alpha #81"
+  },
+  "alpha101_082": {
+   "en": "Kakushadze Alpha #82",
+   "zh": "Kakushadze 因子 Alpha #82"
+  },
+  "alpha101_083": {
+   "en": "Kakushadze Alpha #83",
+   "zh": "Kakushadze 因子 Alpha #83"
+  },
+  "alpha101_084": {
+   "en": "Kakushadze Alpha #84",
+   "zh": "Kakushadze 因子 Alpha #84"
+  },
+  "alpha101_085": {
+   "en": "Kakushadze Alpha #85",
+   "zh": "Kakushadze 因子 Alpha #85"
+  },
+  "alpha101_086": {
+   "en": "Kakushadze Alpha #86",
+   "zh": "Kakushadze 因子 Alpha #86"
+  },
+  "alpha101_087": {
+   "en": "Kakushadze Alpha #87",
+   "zh": "Kakushadze 因子 Alpha #87"
+  },
+  "alpha101_088": {
+   "en": "Kakushadze Alpha #88",
+   "zh": "Kakushadze 因子 Alpha #88"
+  },
+  "alpha101_089": {
+   "en": "Kakushadze Alpha #89",
+   "zh": "Kakushadze 因子 Alpha #89"
+  },
+  "alpha101_090": {
+   "en": "Kakushadze Alpha #90",
+   "zh": "Kakushadze 因子 Alpha #90"
+  },
+  "alpha101_091": {
+   "en": "Kakushadze Alpha #91",
+   "zh": "Kakushadze 因子 Alpha #91"
+  },
+  "alpha101_092": {
+   "en": "Kakushadze Alpha #92",
+   "zh": "Kakushadze 因子 Alpha #92"
+  },
+  "alpha101_093": {
+   "en": "Kakushadze Alpha #93",
+   "zh": "Kakushadze 因子 Alpha #93"
+  },
+  "alpha101_094": {
+   "en": "Kakushadze Alpha #94",
+   "zh": "Kakushadze 因子 Alpha #94"
+  },
+  "alpha101_095": {
+   "en": "Kakushadze Alpha #95",
+   "zh": "Kakushadze 因子 Alpha #95"
+  },
+  "alpha101_096": {
+   "en": "Kakushadze Alpha #96",
+   "zh": "Kakushadze 因子 Alpha #96"
+  },
+  "alpha101_097": {
+   "en": "Kakushadze Alpha #97",
+   "zh": "Kakushadze 因子 Alpha #97"
+  },
+  "alpha101_098": {
+   "en": "Kakushadze Alpha #98",
+   "zh": "Kakushadze 因子 Alpha #98"
+  },
+  "alpha101_099": {
+   "en": "Kakushadze Alpha #99",
+   "zh": "Kakushadze 因子 Alpha #99"
+  },
+  "alpha101_100": {
+   "en": "Kakushadze Alpha #100",
+   "zh": "Kakushadze 因子 Alpha #100"
+  },
+  "alpha101_101": {
+   "en": "Kakushadze Alpha #101",
+   "zh": "Kakushadze 因子 Alpha #101"
+  },
+  "fund_asset_growth": {
+   "en": "Asset growth - inverse investment factor",
+   "zh": "资产增长 - 反向投资因子"
+  },
+  "fund_earnings_yield": {
+   "en": "Earnings yield - net income over market cap",
+   "zh": "盈利收益率 - 净利润除以市值"
+  },
+  "fund_gross_profitability": {
+   "en": "Gross profitability - gross profit over total assets",
+   "zh": "毛利盈利能力 - 毛利除以总资产"
+  },
+  "fund_roe": {
+   "en": "ROE - return on equity (PIT-safe fundamentals)",
+   "zh": "ROE - 净资产收益率，基于时点正确（PIT）基本面"
+  }
+ },
+ "themes": {
+  "volume": "成交量",
+  "momentum": "动量",
+  "reversal": "反转",
+  "volatility": "波动率",
+  "microstructure": "微观结构",
+  "quality": "质量",
+  "liquidity": "流动性",
+  "value": "价值",
+  "growth": "成长",
+  "sentiment": "情绪"
+ }
+}
diff --git a/wiki/alpha-library/index.html b/wiki/alpha-library/index.html
index d6864ce3..d02bdb4a 100644
--- a/wiki/alpha-library/index.html
+++ b/wiki/alpha-library/index.html
@@ -24,11 +24,11 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
-  <script type="module" src="/alpha-library/alpha.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
+  <script type="module" src="/alpha-library/alpha.js?v=20260929e"></script>
   <style>
     .alpha-hero { width: min(1180px, calc(100% - 32px)); margin: 56px auto 32px; }
     .alpha-hero h1 { font-family: var(--serif); font-size: clamp(2.2rem, 4vw, 3.4rem); margin: 8px 0 14px; letter-spacing: -0.01em; }
diff --git a/wiki/docs/content.js b/wiki/docs/content.js
index 7fbeaa8b..5845e832 100644
--- a/wiki/docs/content.js
+++ b/wiki/docs/content.js
@@ -1,5 +1,5 @@
-import { DOCS_STRUCTURE as EN_STRUCTURE } from "/docs/content.en.js?v=20260929d";
-import { DOCS_STRUCTURE as ZH_STRUCTURE } from "/docs/content.zh.js?v=20260929d";
+import { DOCS_STRUCTURE as EN_STRUCTURE } from "/docs/content.en.js?v=20260929e";
+import { DOCS_STRUCTURE as ZH_STRUCTURE } from "/docs/content.zh.js?v=20260929e";
 
 // One content set per language; every version alias renders the latest.
 export const DOCS_DEFAULT_VERSION = "0.1.16";
diff --git a/wiki/docs/index.html b/wiki/docs/index.html
index ca883eea..603136a0 100644
--- a/wiki/docs/index.html
+++ b/wiki/docs/index.html
@@ -21,11 +21,11 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <link rel="stylesheet" href="/docs/styles.css?v=20260929d">
-  <script type="module" src="/docs/main.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/docs/styles.css?v=20260929e">
+  <script type="module" src="/docs/main.js?v=20260929e"></script>
 </head>
 <body class="docs-body" data-section="docs">
   <!-- site-header:start -->
diff --git a/wiki/docs/main.js b/wiki/docs/main.js
index c4a6c944..281f194f 100644
--- a/wiki/docs/main.js
+++ b/wiki/docs/main.js
@@ -1,11 +1,11 @@
-import { initSite, setPageLang, storedLang } from "/site.js?v=20260929d";
+import { initSite, setPageLang, storedLang } from "/site.js?v=20260929e";
 import {
   DOCS_DEFAULT_LANG,
   DOCS_DEFAULT_PAGE,
   DOCS_LANGUAGES,
   DOCS_UI,
   DOCS_VERSIONS
-} from "/docs/content.js?v=20260929d";
+} from "/docs/content.js?v=20260929e";
 
 const SITE = "https://vibetrading.wiki";
 
diff --git a/wiki/home/index.html b/wiki/home/index.html
index fba0bfca..22a53d1a 100644
--- a/wiki/home/index.html
+++ b/wiki/home/index.html
@@ -23,10 +23,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
 </head>
 <body data-section="home">
   <!-- site-header:start -->
diff --git a/wiki/locales/zh.json b/wiki/locales/zh.json
index 9c24faf8..fe8a6de4 100644
--- a/wiki/locales/zh.json
+++ b/wiki/locales/zh.json
@@ -105,5 +105,6 @@
   "home.term.route": "路由：加密行情 -> 策略 -> 回测 -> 运行卡",
   "home.term.result": "<span class=\"ok\">收益</span> +18.6% <span class=\"warn\">最大回撤</span> -7.4%",
   "home.term.workers": "成员：宏观、量化、风控、催化剂",
-  "home.term.report": "<span class=\"ok\">报告</span>已保存，附证据、指标和注意事项"
+  "home.term.report": "<span class=\"ok\">报告</span>已保存，附证据、指标和注意事项",
+  "lab.gtja.date": "2026-05-17 · 2026-09-29 修订"
 }
diff --git a/wiki/main.js b/wiki/main.js
index 19e30d0f..4bf28027 100644
--- a/wiki/main.js
+++ b/wiki/main.js
@@ -1,4 +1,4 @@
-import { initSite } from "/site.js?v=20260929d";
+import { initSite } from "/site.js?v=20260929e";
 
 // Home page install tabs; a no-op on pages without them.
 function initInstallTabs() {
diff --git a/wiki/partials/site-head.html b/wiki/partials/site-head.html
index 9e9df5fd..46668abb 100644
--- a/wiki/partials/site-head.html
+++ b/wiki/partials/site-head.html
@@ -2,5 +2,5 @@
 <link rel="preconnect" href="https://fonts.googleapis.com">
 <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
 <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-<link rel="stylesheet" href="/styles.css?v=20260929d">
-<script src="/theme-init.js?v=20260929d"></script>
+<link rel="stylesheet" href="/styles.css?v=20260929e">
+<script src="/theme-init.js?v=20260929e"></script>
diff --git a/wiki/research-lab/index.html b/wiki/research-lab/index.html
index 8fdd313c..e97eb9fd 100644
--- a/wiki/research-lab/index.html
+++ b/wiki/research-lab/index.html
@@ -18,10 +18,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
 </head>
 <body data-section="lab">
   <!-- site-header:start -->
@@ -62,7 +62,7 @@
 
     <section class="posts-grid" aria-label="Latest posts" data-i18n-attr="aria-label=lab.postsLabel">
       <article class="post-card">
-        <p class="post-card__date">2026-05-17</p>
+        <p class="post-card__date" data-i18n="lab.gtja.date">2026-05-17 · revised 2026-09-29</p>
         <h2 class="post-card__title">
           <a href="/research-lab/posts/alpha-191-in-2026.html" data-href-zh="/research-lab/posts/alpha-191-in-2026-zh.html" data-i18n="lab.gtja.title">Which of the 191 GTJA alphas still work in 2026?</a>
         </h2>
diff --git a/wiki/research-lab/posts/alpha-191-in-2026-zh.html b/wiki/research-lab/posts/alpha-191-in-2026-zh.html
index 66892937..0b0cde8a 100644
--- a/wiki/research-lab/posts/alpha-191-in-2026-zh.html
+++ b/wiki/research-lab/posts/alpha-191-in-2026-zh.html
@@ -22,10 +22,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
   <style>
     .post {
       width: min(760px, calc(100% - 32px));
@@ -255,37 +255,39 @@
         <p class="lede">2014 年，国泰君安证券发布了一份研究报告，系统整理了面向 A 股市场的 191 个短周期 alpha 因子。十二年过去，T+1 交收制度依旧，行业轮动格局已两度翻转，散户资金流处于数十年来的高位&mdash;&mdash;这些公式里还有多少能产生可靠的信号？我们在沪深300 上、以 2018&ndash;2025 年为区间，对全部 191 个因子做了基准测试，得到的答案远比一个数字更耐人寻味。</p>
       </header>
 
+      <p class="callout" id="revision"><strong>2026-09-29 修订。</strong>本文第一版（2026-05-17）使用的 Tushare 价格未做除权除息复权&mdash;&mdash;跨越除权日的收益率读到的是机械性的价格缺口，而且总是偏向下跌&mdash;&mdash;并且把最新的沪深300 成分名单套用到了整个窗口。这两个问题已于 2026-08-04 修复；自 2026-09 起，某根 K 线上只要因子的任一输入缺失，因子就不输出值，而不是用缺失数据生成信号。第一版报告的是存活 10 个、反转 15 个、失效 165 个；下文所有数字均来自在当前代码上的重新运行。</p>
+
       <section id="tldr">
         <h2>要点速览</h2>
-        <p>国泰君安 2014 年发布的 191 个短周期 alpha 因子中，在沪深300、2018&ndash;2025 年区间上，只有 10 个（5%）仍能通过我们的存活筛选&mdash;&mdash;平均 IC 为正且高于 0.02、t 统计量高于 2、且至少 55% 的交易日 IC 为正。另有 15 个（8%）<em>方向反转</em>且具有统计显著性，如今成了反向信号。其余 165 个（87%）要么已衰减到不再显著，要么因缺少我们没有的行业标签而无法计算，要么触发了 NaN 占比 &gt;95% 的预热期保护。这 10 个存活因子集中在微观结构 / K 线形态一族，而不是我们原先预期的原始量价交互一族。具体数量见下文；方法刻意保守，局限性也不容忽视。</p>
+        <p>国泰君安 2014 年发布的 191 个短周期 alpha 因子中，在沪深300、2018&ndash;2025 年区间上，只有 8 个（4%）仍能通过我们的存活筛选&mdash;&mdash;平均 IC 为正且高于 0.02、t 统计量高于 2、且至少 55% 的交易日 IC 为正。另有 14 个（7%）<em>方向反转</em>且具有统计显著性，如今成了反向信号。另外 168 个（88%）在两个方向上都已不再显著；还有 1 个（<code>gtja191_138</code>）完全无法检验：它的输出 99% 是 NaN。8 个存活因子中有一半带有微观结构 / K 线形态标签，这是各主题中最高的存活率，而不是我们原先预期的原始量价交互一族。具体数量见下文；方法刻意保守，局限性也不容忽视。</p>
         <div class="findings-stats" aria-label="因子存活数量汇总">
           <div class="stat alive">
             <span class="label">存活</span>
-            <span class="num"><span class="num-alive">10</span></span>
+            <span class="num"><span class="num-alive">8</span></span>
             <span>191 个因子中</span>
           </div>
           <div class="stat reversed">
             <span class="label">反转</span>
-            <span class="num"><span class="num-reversed">15</span></span>
+            <span class="num"><span class="num-reversed">14</span></span>
             <span>191 个因子中</span>
           </div>
           <div class="stat dead">
             <span class="label">失效</span>
-            <span class="num"><span class="num-dead">165</span></span>
+            <span class="num"><span class="num-dead">168</span></span>
             <span>191 个因子中</span>
           </div>
         </div>
-        <p class="callout">下文数字均为对内置 gtja191 因子集实际运行 W4.a bench 的输出（沪深300，2018&ndash;2025），可用文末的 CLI 命令复现。</p>
+        <p class="callout">下文数字来自 2026-09-29 的一次 bench 运行（Vibe-Trading 0.1.16：沪深300 时点成分、复权价格，2018&ndash;2025），可用文末的 CLI 命令复现。</p>
       </section>
 
       <section id="background">
         <h2>背景</h2>
-        <p>国泰君安 2014 年的这份研究报告&mdash;&mdash;题为《191 个短周期交易型 alpha 因子》（英文译名 <em>191 Short-Period Transactional Alpha Factors</em>）&mdash;&mdash;问世于中国量化发展的一个特殊时点。当时市场刚走出多年的横盘震荡，散户参与度正在回升，而 Kakushadze 式的“公式化 alpha”模板在此前几年已借由他 2015 年的 arXiv 预印本在国际上流行开来。回头看，国泰君安团队拿出的是迄今针对 A 股公开发布的最系统的短周期因子目录之一：191 个编号公式，每个都只是基于日频 OHLCV 与换手率的寥寥几行算子运算。</p>
+        <p>国泰君安 2014 年的这份研究报告&mdash;&mdash;题为《191 个短周期交易型 alpha 因子》（英文译名 <em>191 Short-Period Transactional Alpha Factors</em>）&mdash;&mdash;问世于中国量化发展的一个特殊时点。当时市场刚走出多年的横盘震荡，散户参与度正在回升，而“公式化 alpha”模板要到一年后，才借由 Kakushadze 2015 年的 arXiv 预印本 <em>101 Formulaic Alphas</em> 在国际上流行开来。回头看，国泰君安团队拿出的是迄今针对 A 股公开发布的最系统的短周期因子目录之一：191 个编号公式，每个都只是基于日频 OHLCV 与换手率的寥寥几行算子运算。</p>
         <p>报告中的公式读起来像是关于市场微观结构的第一性原理假设，被翻译成了便于用 pandas 实现的算术。有些事后看来显而易见（基于排名的五日反转；成交量与收盘价的相关性）；另一些则相当奇特（报告中
         <code class="formula">SUMIF</code> / <code class="formula">FILTER</code> / <code class="formula">REGBETA</code>
         的组合嵌套深达四到五层算子）。用现代量化的话说，它们是一个<em>先验库</em>：每个公式都编码了某个人的判断&mdash;&mdash;哪种微观结构规律可以在 1&ndash;5 天的周期上被利用。</p>
         <p>这件事在 2026 年之所以重要，原因很简单：短周期 alpha 的衰减速度比任何其他类别都快。长周期的价值或质量因子不做任何调整也可能存活十年；而建立在量价交互上的 1&ndash;5 天公式大概做不到。在因子研究的尺度上，十二年的样本外几乎等同于永恒。A 股市场本身也变了：T+1 交收仍在，但机构与散户的结构已经翻转，科创板与创业板注册制改革重新定价了小盘股风险，公募基金带有高频色彩的交易执行也压缩了许多明显的均值回归窗口。</p>
-        <p>于是我们做了这次检验。下面先介绍方法，数字随后给出。</p>
+        <p>于是我们做了这次检验：先讲方法，再给数字。</p>
       </section>
 
       <section id="method">
@@ -293,61 +295,61 @@
         <p>这次检验刻意做成<em>截面信息系数（IC）研究</em>，而不是完整回测。要回答的问题是：<em>这个 alpha 对股票的排序，是否在整个窗口内平均而言、稳健地与次日收益相关？</em>更高保真度的策略回测（含交易成本、仓位限制、行业中性化、按衰减系数加权的组合）是另一个问题，不在本文讨论范围内。</p>
 
         <h3>股票池</h3>
-        <p>沪深300。流动性最好的 300 只 A 股，由指数公司每半年调整一次成分。我们在每个调仓日使用指数的现行成分股，因此股票池随指数漂移。相对于固定不变的股票池，这会带来偏差（被调出沪深300 的恰恰是流动性已经衰退的个股，这可能使 IC 估计产生偏差），但它与实际部署时使用该指数的方式一致。</p>
+        <p>沪深300：流动性最好的 300 只 A 股，由指数公司每半年调整一次成分。每个交易日的截面，取 Tushare 月末成分名单中当日有效的指数成分，因此一只股票只在它属于指数的期间计入&mdash;&mdash;窗口期内共有 348 只股票曾是成分股。价格已做除权除息前复权。</p>
 
         <h3>区间</h3>
         <p>2018-01-02 至 2025-12-31。八个日历年，约 1,940 个交易日，相对 2014 年的报告完全处于样本外。窗口刻意覆盖了 2018 年的熊市、2020&ndash;2021 年流动性驱动的上涨、2022 年的回撤以及 2024&ndash;2025 年的震荡市&mdash;&mdash;因此平均 IC 是跨市场状态平均的结果，而不是挑选有利市场状态得出的。</p>
 
         <h3>信号定义</h3>
-        <p>对每个 alpha 和每个交易日 <em>t</em>，我们计算股票池内每只股票的因子值，再在截面上做排名变换，映射到 <code class="formula">[0, 1]</code>。远期收益为从 <em>t</em> 日收盘到 <em>t+1</em> 日收盘的 1 日对数收益率，同样做截面排名变换。第 <em>t</em> 日的 Spearman IC 即两条排名序列的 Pearson 相关系数。我们报告所有有效交易日的平均 IC、IC 序列的 t 统计量，以及 IC 为正的交易日占比。</p>
+        <p>对每个 alpha 和每个交易日 <em>t</em>，我们计算股票池内每只股票的因子值，再在截面上做排名变换，映射到 <code class="formula">[0, 1]</code>。远期收益为从 <em>t</em> 日收盘到 <em>t+1</em> 日收盘的 1 日简单收益率，同样做截面排名变换。第 <em>t</em> 日的 Spearman IC 即两条排名序列的 Pearson 相关系数。我们报告所有有效交易日的平均 IC、IC 序列的 t 统计量，以及 IC 为正的交易日占比。</p>
 
         <h3>分类标准</h3>
         <ul>
           <li><strong>存活。</strong>平均 IC &gt; 0.02、t-stat(IC) &gt; 2，且 IC &gt; 0 的交易日占比 &ge; 55%。三个条件须同时满足。</li>
           <li><strong>反转。</strong>平均 IC &lt; &minus;0.02 且 t-stat(IC) &lt; &minus;2（即上述条件的符号翻转版）。原报告设定该因子以某一方向预测收益；我们观察到它以相反方向预测，且具有统计显著性。</li>
-          <li><strong>失效。</strong>其余所有情况：|平均 IC| &le; 0.02，或 |t-stat| &le; 2，或该因子无法在窗口内干净地计算（例如需要我们没有的日内逐笔数据）。</li>
+          <li><strong>失效。</strong>其余所有能够计算的情况：|平均 IC| &le; 0.02，或 |t-stat| &le; 2，或 IC 为正的交易日不足 55%。输出中 NaN 占比超过 95% 的因子完全不参与检验，单独列出。</li>
         </ul>
 
         <h3>方法本身带来的局限</h3>
         <p>以下三个刻意的选择限定了结果应当如何解读：</p>
         <ul>
           <li>IC 为<em>向前 1 日</em>，不做衰减平滑。decay-3 或 decay-5 IC（原报告偏好的周期）会得出略有不同的数字；我们会在后续文章中发布。</li>
-          <li>不做交易成本调整。一个高 IC、每日调仓换手率接近 100% 的因子，在扣除 5&ndash;10 bps 的现实双边成本后，实际上无法盈利。我们在 bench 产出物中把换手率与 IC 一并报告，但这里的存活分类没有考虑它。</li>
-          <li>不做行业中性化。报告中部分因子依赖行业映射（<code class="formula">indneutralize</code>）；对这些因子，我们退而用每日截面均值替代，这是一种严格更弱的中性化。受影响的因子在逐因子输出中有标注。</li>
+          <li>不做交易成本调整。一个高 IC、每日调仓换手率接近 100% 的因子，在扣除 5&ndash;10 bps 的现实双边成本后，实际上无法盈利。bench 不测量换手率，这里的存活分类也没有考虑它。</li>
+          <li>不做行业中性化。国泰君安 191 因子没有一个需要行业标签，因此这里所有 IC 都是在原始截面上测得的。</li>
         </ul>
 
         <h3>所用工具</h3>
         <pre><code>$ vibe-trading alpha bench --zoo gtja191 --universe csi300 \
                             --period 2018-2025 --top 20</code></pre>
-        <p>一条命令即可。输出为一份 HTML 报告，包含逐因子的 IC、t-stat、换手率和 IC 时间序列图。复现方法见文末。</p>
+        <p>一条命令即可。它会在 <code>~/.vibe-trading/reports/</code> 下写一份 HTML 报告，列出按 IR 排名靠前的因子（平均 IC、IC 标准差、IR、IC 为正的交易日占比、交易日数），以及无法计算的因子。复现方法见文末。</p>
       </section>
 
       <section id="findings">
         <h2>结果</h2>
 
         <h3>总体存活情况</h3>
-        <p>再列一次核心数字，附带同样的说明：最终数字在 W4.a bench 跑完后落定：</p>
+        <p>再列一次核心数字：</p>
         <div class="findings-stats" aria-label="因子存活数量汇总（结果部分重复展示）">
           <div class="stat alive">
             <span class="label">存活</span>
-            <span class="num"><span class="num-alive">10</span></span>
+            <span class="num"><span class="num-alive">8</span></span>
             <span>通过全部三项筛选</span>
           </div>
           <div class="stat reversed">
             <span class="label">反转</span>
-            <span class="num"><span class="num-reversed">15</span></span>
+            <span class="num"><span class="num-reversed">14</span></span>
             <span>如今以相反方向预测</span>
           </div>
           <div class="stat dead">
             <span class="label">失效</span>
-            <span class="num"><span class="num-dead">165</span></span>
+            <span class="num"><span class="num-dead">168</span></span>
             <span>低于显著性阈值</span>
           </div>
         </div>
-        <p>在早期的部分运行中，让我们意外的不是失效数量&mdash;&mdash;衰减人人都预料得到&mdash;&mdash;而是<em>反转</em>数量。相当一部分在 2014 年有效的因子，如今成了幅度具有统计显著性的反向指标。最简单的解读是：公式原本捕捉的行为异象（小盘股均值回归、散户的尾盘资金流）已被恰恰是这类公式所代表的系统化交易挤出，剩下的就是反向交易。</p>
+        <p>让我们意外的不是失效数量&mdash;&mdash;衰减人人都预料得到&mdash;&mdash;而是<em>反转</em>数量。相当一部分在 2014 年有效的因子，如今成了幅度具有统计显著性的反向指标。最简单的解读是：公式原本捕捉的行为异象（小盘股均值回归、散户的尾盘资金流）已被恰恰是这类公式所代表的系统化交易挤出，剩下的就是反向交易。</p>
 
         <h3>分主题拆解</h3>
-        <p>我们按主导主题&mdash;&mdash;即因子最倚重的那组算子&mdash;&mdash;给每个 alpha 手工打了标签。这是一种有损的分类（许多因子融合了两个主题），但总体规律对标签选择是稳健的。下表数字为各主题内的存活率：</p>
+        <p>每个因子在注册表元数据里带有一到两个主题标签&mdash;&mdash;即它最倚重的那组算子。带两个标签的因子在两行中都会计入，因此各行数量之和超过 190。下表数字为各主题内的存活率：</p>
 
         <table class="theme-table" aria-label="各主题存活率">
           <thead>
@@ -363,7 +365,7 @@
               <td>量价交互</td>
               <td>成交量与收盘价、高低价振幅的相关性 / 协方差</td>
               <td>81</td>
-              <td>5% (4/81)</td>
+              <td>4% (3/81)</td>
             </tr>
             <tr>
               <td>短周期波动率</td>
@@ -375,13 +377,13 @@
               <td>反转</td>
               <td>1-5 日周期上取负号的收益信号</td>
               <td>38</td>
-              <td>11% (4/38)</td>
+              <td>8% (3/38)</td>
             </tr>
             <tr>
               <td>动量</td>
               <td>10-60 日周期上取正号的收益信号</td>
               <td>63</td>
-              <td>2% (1/63)</td>
+              <td>0% (0/63)</td>
             </tr>
             <tr>
               <td>换手 / 流动性</td>
@@ -395,61 +397,67 @@
               <td>18</td>
               <td>22% (4/18)</td>
             </tr>
+            <tr>
+              <td>情绪</td>
+              <td>市场下跌日中个股上涨的占比（仅 <code>gtja191_075</code> 一个因子，基准为代理变量）</td>
+              <td>1</td>
+              <td>0% (0/1)</td>
+            </tr>
           </tbody>
         </table>
 
-        <p>W4.a bench 的实际结果是：<em>微观结构 / 振幅</em>类因子是最突出的存活者（存活率 22%，18 个中 4 个），<em>反转</em>类次之（11%，38 个中 4 个）。我们原本预期最经得起时间考验的类别&mdash;&mdash;原始<em>量价交互</em>（5%）和<em>短周期波动率</em>（8%）&mdash;&mdash;衰减程度反而超过开收高低价分解一族。<em>动量</em>类 2%（63 个中 1 个）和规模很小的<em>换手</em>类 0%，是被套利得最干净的两组。这一规律符合如下情形下的预期：十年的系统化资金压缩了最容易的反转 / 动量交易，却没有触动那些捕捉 K 线形态效应、基于日线结构几何的因子。</p>
+        <p>重新运行的结果是：<em>微观结构 / 振幅</em>类因子是最突出的存活者（22%，18 个中 4 个）。<em>反转</em>和<em>短周期波动率</em>其次，都是 8%（38 个中 3 个、26 个中 2 个），原始<em>量价交互</em>为 4%（81 个中 3 个）。<em>动量</em>类 63 个无一存活&mdash;&mdash;其中 5 个如今以相反方向预测&mdash;&mdash;规模很小的<em>换手</em>类 2 个也无一存活。这一规律符合如下情形下的预期：十年的系统化资金压缩了最容易的反转 / 动量交易，却没有触动那些捕捉 K 线形态效应、基于日线结构几何的因子。</p>
 
         <h3>存活因子 Top 5</h3>
-        <p>窗口内平均 IC 最高的五个因子，按降序排列。每个因子以其 zoo id 和公式的转述标识；完整的算子级定义见对应 Python 模块中的 <code>__alpha_meta__["formula_latex"]</code> 字段。</p>
+        <p>窗口内 IR（平均 IC 除以其标准差）最高的五个因子，按降序排列，公式均为注册表中的原样（对应 Python 模块的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
 
-                <article class="alpha-card">
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_114</span>
+          <pre class="formula-block"><code>((RANK(DELAY(((HIGH-LOW)/(SUM(CLOSE,5)/5)),2)) * RANK(RANK(VOLUME))) / (((HIGH-LOW)/(SUM(CLOSE,5)/5))/(VWAP-CLOSE)))</code></pre>
+          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0606，IR = 0.2926。公式逐字取自注册表（<code>gtja191_114</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
+        </article>
+        <article class="alpha-card">
           <span class="alpha-id">gtja191_171</span>
           <pre class="formula-block"><code>-1*((l-c)*(o^5))/((c-h)*(c^5))</code></pre>
-          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0432，IR = 0.2690。公式逐字取自注册表（<code>gtja191_171</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
+          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0435，IR = 0.2299。公式逐字取自注册表（<code>gtja191_171</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
+        </article>
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_120</span>
+          <pre class="formula-block"><code>rank(vwap-close)/rank(vwap+close)</code></pre>
+          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0452，IR = 0.2119。公式逐字取自注册表（<code>gtja191_120</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_111</span>
           <pre class="formula-block"><code>sma(v*((c-l)-(h-c))/(h-l),11,2)-sma(v*((c-l)-(h-c))/(h-l),4,2)</code></pre>
-          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0349，IR = 0.2232。公式逐字取自注册表（<code>gtja191_111</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
+          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0355，IR = 0.1888。公式逐字取自注册表（<code>gtja191_111</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_163</span>
           <pre class="formula-block"><code>rank(((-1*ret)*mean(v,20))*vwap*(high-close))</code></pre>
-          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0347，IR = 0.2008。公式逐字取自注册表（<code>gtja191_163</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_002</span>
-          <pre class="formula-block"><code>(-1 * DELTA(((CLOSE - LOW) - (HIGH - CLOSE)) / (HIGH - LOW), 1))</code></pre>
-          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0262，IR = 0.1619。公式逐字取自注册表（<code>gtja191_002</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_054</span>
-          <pre class="formula-block"><code>((-1*RANK((STD(ABS(CLOSE-OPEN),10)+(CLOSE-OPEN))+CORR(CLOSE,OPEN,10))))</code></pre>
-          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0272，IR = 0.1606。公式逐字取自注册表（<code>gtja191_054</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
+          <p class="alpha-note">在沪深300 / 2018&ndash;2025 窗口上，平均 IC = 0.0342，IR = 0.1674。公式逐字取自注册表（<code>gtja191_163</code> 的 <code>__alpha_meta__["formula_latex"]</code>）。</p>
         </article>
 
-        <h3>三个著名的失效因子</h3>
-        <p>以下三个公式在 2014 年的报告中有效（或被声称有效），如今却稳稳落在失效或反转一栏：</p>
+        <h3>三个如今方向相反的因子</h3>
+        <p>平均 IC 最负的三个因子。三者都在反转一栏：它们如今以与报告设定相反的方向预测收益，且具有统计显著性。</p>
 
-                <article class="alpha-card">
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_124（反转）</span>
+          <pre class="formula-block"><code>(close-vwap)/decay_linear(rank(tsmax(close,30)),2)</code></pre>
+          <p class="alpha-note">平均 IC = -0.0518，IR = -0.2516。在沪深300 / 2018&ndash;2025 上，是 gtja191 因子集中平均 IC 最负的因子。</p>
+        </article>
+        <article class="alpha-card">
           <span class="alpha-id">gtja191_178（反转）</span>
           <pre class="formula-block"><code>(c-delay(c,1))/delay(c,1)*v</code></pre>
-          <p class="alpha-note">平均 IC = -0.0327，IR = -0.1930。按原始 IC 计，属于 gtja191 因子集在沪深300 / 2018&ndash;2025 上表现最差的一批。</p>
+          <p class="alpha-note">平均 IC = -0.0329，IR = -0.1644。平均 IC 第二负。</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_137（反转）</span>
-          <pre class="formula-block"><code>见正文</code></pre>
-          <p class="alpha-note">平均 IC = -0.0277，IR = -0.1556。按原始 IC 计，属于 gtja191 因子集在沪深300 / 2018&ndash;2025 上表现最差的一批。</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_066（反转）</span>
-          <pre class="formula-block"><code>(CLOSE-MEAN(CLOSE,6))/MEAN(CLOSE,6)*100</code></pre>
-          <p class="alpha-note">平均 IC = -0.0270，IR = -0.1377。按原始 IC 计，属于 gtja191 因子集在沪深300 / 2018&ndash;2025 上表现最差的一批。</p>
+          <pre class="formula-block"><code>16*((c-dc1+(c-o)/2+dc1-do1)/MAX_term) * MAX(abs(h-dc1), abs(l-dc1))</code></pre>
+          <p class="alpha-note">平均 IC = -0.0300，IR = -0.1464。平均 IC 第三负。<code>MAX_term</code> 是报告中分三段取值的分母，由 |h &minus; dc1|、|l &minus; dc1|、|h &minus; dl1| 和 |dc1 &minus; do1| 构成（dc1、do1、dl1 分别为前一日的收盘价、开盘价和最低价）；完整定义见模块代码。</p>
         </article>
 
         <h3>一段反思</h3>
-        <p>做完这类存活研究，人们容易过度概括：“衰减不可避免；公式化 alpha 已死。”我们认为这不是正确的解读。这次研究给出的结论更窄、也更有用：一份 12 年前的短周期因子目录中，仍有相当一部分能产生信号；存活者集中在可解释的主题里（量价交互、短周期波动率），失效者也同样集中在可解释的主题里（朴素反转、简单换手率变换）。至于 Kakushadze 101 公式或 Qlib 158 特征集是否会以同样的速度衰减，它几乎说明不了什么&mdash;&mdash;这两个因子集的算子词汇不同，面向的股票池也不同（分别是美股和多周期）。这留待后续工作，另行跑 bench。</p>
+        <p>做完这类存活研究，人们容易过度概括：“衰减不可避免；公式化 alpha 已死。”我们认为这不是正确的解读。这次研究给出的结论更窄、也更有用：一份 12 年前的短周期因子目录中，仍有一小部分&mdash;&mdash;190 个中的 8 个&mdash;&mdash;能产生信号；存活者集中在一个可解释的主题里（K 线几何：开收高低价分解），失效者也同样集中在可解释的主题里（动量类无一存活、5 个方向翻转，以及简单的换手率变换）。至于 Kakushadze 101 公式或 Qlib 158 特征集是否会以同样的速度衰减，它几乎说明不了什么&mdash;&mdash;这两个因子集的算子词汇不同，面向的股票池也不同（分别是美股和多周期）。这留待后续工作，另行跑 bench。</p>
       </section>
 
       <section id="caveats">
@@ -463,13 +471,16 @@
         <p>我们的基准测试只覆盖流动性最好的 300 只 A 股。为全 A 股票池设计的因子&mdash;&mdash;全 A 约有 5,000 只股票，流动性特征差异很大&mdash;&mdash;表现会有所不同。尤其是小盘股均值回归类因子，在沪深300 上往往比在全市场上表现更差（因为沪深300 以机构资金为主导，短周期的散户反转效应较弱），这里一些“失效”的因子在中证1000 上或许会复活。</p>
 
         <h3>Tushare 数据范围</h3>
-        <p>我们的数据源（Tushare 日终 OHLCV 与换手率，外加由成交额推算的 VWAP）不含日内逐笔或订单簿信息。少数国泰君安 191 因子在原始表述中引用了日内数据（深度加权 VWAP、level-1 盘口失衡）；对这些因子，我们以日频 VWAP 替代，并在逐因子输出中标注这一替换。这些因子在这里应视为<em>未经充分检验</em>，而不是“失效”。</p>
+        <p>我们的数据源是 Tushare 日终 OHLCV 与成交额，已做除权除息前复权，VWAP 由成交额推算；不含日内逐笔或订单簿数据。有 5 个国泰君安 191 因子引用了面板中没有的输入&mdash;&mdash;基准指数收盘价（<code>gtja191_075</code>、<code>_149</code>、<code>_181</code>、<code>_182</code>）或多因子回归残差（<code>gtja191_030</code>）&mdash;&mdash;改用了代理变量：截面平均收盘价，或日收益率平方的加权平均，各因子的说明中均有写明。这 5 个因子在这里应视为<em>未经充分检验</em>，而不是“失效”。</p>
 
         <h3>8 年窗口偏短</h3>
         <p>按学术标准，8 年窗口偏短。经典的 Fama-French 论文使用 30&ndash;90 年的窗口。一些被我们标为“失效”的因子，在更长或所处时段不同的样本上可能复活&mdash;&mdash;例如，若 2026&ndash;2030 年的市场状态重新回到散户占比更高的结构，均值回归类因子可能恢复。这个标签的含义是“在本窗口内失效”，而非“永久失效”。</p>
 
         <h3>股票池中的幸存者偏差</h3>
-        <p>本文使用的沪深300 成分股名单，是把<em>当前</em>的指数成分套用到整个 2018&ndash;2025 窗口，而不是按每次调仓时点重建的指数成分（point-in-time）。窗口期内退市、降档或被调出的股票不在其中；后期才纳入的股票则被回填了完整的历史价格。这是一种标准的幸存者偏差，会使 IC 估计在纯多头方向上偏高。对于以同样方式构建美股股票池的任何跨因子集比较，这一提醒更加适用&mdash;&mdash;例如，用今天的成分股名单跑标普 500，相对于真正的时点股票池会高估存活数量；那里若得出 0% 的存活比例，应理解为“衰减加上幸存者偏差把信号抵消了”，而不是该因子家族的干净失败。时点指数成分是计划中的升级。</p>
+        <p>本文第一版把最新的沪深300 成分名单套用到整个 2018&ndash;2025 窗口，每只股票之所以在列，都是因为它存活到了最后&mdash;&mdash;这是一个事后挑选的集合，会使 IC 偏高。重新运行时，每个交易日的截面都按当日有效的成分名单筛选（窗口期内共有 348 只股票曾是成分股）。<code>sp500</code> 股票池仍使用今天的成分名单，因此在它上面做的任何跨因子集比较都适用同样的提醒&mdash;&mdash;它的 bench 报告会写明这一点&mdash;&mdash;那里较低的存活数应理解为“衰减加上幸存者偏差”，而不是该因子家族的干净失败。</p>
+
+        <h3>190 个公式，各测一次</h3>
+        <p>在同一窗口上检验 190 个公式，总会有一些仅凭运气显得不错。bench 会报告：如果所有公式都不含任何信号，预期的最高 IR 有多大&mdash;&mdash;这里是 0.204。实际最高的 IR 为 0.293（<code>gtja191_114</code>），通过了这一校正（概率 0.9999）。另有两个存活因子高于 0.204&mdash;&mdash;<code>gtja191_171</code>（0.230）和 <code>gtja191_120</code>（0.212）；其余五个的 IR 在 0.130 到 0.189 之间，低于这一水平，应视为候选信号而非已确认的信号。</p>
 
         <h3>未按市场状态分组</h3>
         <p>我们在整个 2018&ndash;2025 窗口上对 IC 取平均。这些因子中很多几乎肯定依赖市场状态&mdash;&mdash;在趋势市有效、在震荡市失灵，或者反过来。后续文章会按市场状态（牛市 / 熊市 / 震荡市，以 20 日指数动量划分）拆分存活数量，但本文的汇总数字是跨状态平均的。一个考虑市场状态的部署方案，或许能把一半的失效因子保留为条件信号。</p>
@@ -483,10 +494,10 @@
 export TUSHARE_TOKEN=your_token_here
 vibe-trading alpha bench --zoo gtja191 \
                          --universe csi300 \
-                         --period 2020-2025 \
+                         --period 2018-2025 \
                          --top 20</code></pre>
 
-        <p>你会得到一份保存在 <code>~/.vibe-trading/reports/</code> 的 HTML 报告，包含逐因子的 IC、IR、衰减曲线和一张可排序的 top-N 表。同一命令也适用于 <code>--zoo alpha101</code>（Kakushadze 的 101 Formulaic Alphas，忠实于论文、据 2015 年 arXiv 预印本重写）和 <code>--zoo qlib158</code>（微软 Qlib 特征库，依 Apache-2.0 许可并注明出处使用）。跨因子集比较通过 <code>alpha compare</code> 运行。</p>
+        <p>你会得到一份保存在 <code>~/.vibe-trading/reports/</code> 的 HTML 报告，列出按 IR 排名靠前的因子，以及无法计算的因子。同一命令也适用于 <code>--zoo alpha101</code>（Kakushadze 的 101 Formulaic Alphas，忠实于论文、据 2015 年 arXiv 预印本重写）和 <code>--zoo qlib158</code>（微软 Qlib 特征库，依 Apache-2.0 许可并注明出处使用）。跨因子集比较通过 <code>alpha compare</code> 运行。</p>
 
         <div class="cta-row">
           <a class="primary" href="https://github.com/HKUDS/Vibe-Trading">在 GitHub 查看源码</a>
@@ -494,7 +505,7 @@ vibe-trading alpha bench --zoo gtja191 \
           <a href="/docs/">阅读文档</a>
         </div>
 
-        <p>如果你发现某个因子的存活分类出乎意料&mdash;&mdash;尤其是我们没有列入前 5 名的存活者，或我们放进了存活一栏的著名失效因子&mdash;&mdash;欢迎附上因子 id 和你的理由提交 issue。社区也欢迎通过 pull request 贡献新的因子集、新的股票池（中证1000、纳斯达克 100、主流加密货币）或新的验证工具，流程遵循
+        <p>如果你发现某个因子的存活分类出乎意料&mdash;&mdash;尤其是我们没有列入前 5 名的存活者，或你原本以为会存活却落入反转一栏的因子&mdash;&mdash;欢迎附上因子 id 和你的理由提交 issue。社区也欢迎通过 pull request 贡献新的因子集、新的股票池（中证1000、纳斯达克 100、主流加密货币）或新的验证工具，流程遵循
         <a href="https://github.com/HKUDS/Vibe-Trading/blob/main/CONTRIBUTING.md">CONTRIBUTING.md</a>
         中的 DCO 规范。</p>
 
diff --git a/wiki/research-lab/posts/alpha-191-in-2026.html b/wiki/research-lab/posts/alpha-191-in-2026.html
index 6718205a..92cb136d 100644
--- a/wiki/research-lab/posts/alpha-191-in-2026.html
+++ b/wiki/research-lab/posts/alpha-191-in-2026.html
@@ -22,10 +22,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
   <style>
     .post {
       width: min(760px, calc(100% - 32px));
@@ -259,34 +259,37 @@
         answer turns out to be more interesting than a single number.</p>
       </header>
 
+      <p class="callout" id="revision"><strong>Revised 2026-09-29.</strong> The first version of this post (2026-05-17) ran on Tushare prices that were not adjusted for splits and dividends &mdash; a return across an ex-date read the mechanical price gap, always downward &mdash; and applied today's CSI 300 roster to the whole window. Both were fixed on 2026-08-04, and since 2026-09 an alpha emits no value on a bar where one of its inputs is missing, instead of a signal built on it. That version reported 10 alive, 15 reversed and 165 dead; every number below comes from a re-run on the current code.</p>
+
       <section id="tldr">
         <h2>TL;DR</h2>
-        <p>Of the 191 short-horizon alphas published by GTJA in 2014, only 10 (5%) still pass our
+        <p>Of the 191 short-horizon alphas published by GTJA in 2014, only 8 (4%) still pass our
         alive filter &mdash; positive mean IC above 0.02, t-stat above 2, and at least 55% of days with
-        positive IC &mdash; on CSI 300 over 2018&ndash;2025. A further 15 (8%) have <em>reversed sign</em>
-        with statistical significance and now act as contrarian signals. The remaining 165 (87%) have
-        decayed below significance, could not be computed without sector tags we do not have, or hit
-        the &gt;95% NaN warmup guard. The 10 survivors cluster in the microstructure / shape-of-the-bar
-        family rather than in the raw volume-price interaction family we expected. Exact counts below;
-        the methodology is deliberately conservative and the caveats are non-trivial.</p>
+        positive IC &mdash; on CSI 300 over 2018&ndash;2025. A further 14 (7%) have <em>reversed sign</em>
+        with statistical significance and now act as contrarian signals. Another 168 (88%) no longer clear
+        the significance filter in either direction, and one (<code>gtja191_138</code>) could not be tested
+        at all: 99% of its output is NaN. Half of the 8 survivors carry the microstructure /
+        shape-of-the-bar tag, the best survival rate of any theme, rather than the raw volume-price
+        interaction family we expected. Exact counts below; the methodology is deliberately conservative
+        and the caveats are non-trivial.</p>
         <div class="findings-stats" aria-label="Aggregate alpha survival counts">
           <div class="stat alive">
             <span class="label">Alive</span>
-            <span class="num"><span class="num-alive">10</span></span>
+            <span class="num"><span class="num-alive">8</span></span>
             <span>of 191 alphas</span>
           </div>
           <div class="stat reversed">
             <span class="label">Reversed</span>
-            <span class="num"><span class="num-reversed">15</span></span>
+            <span class="num"><span class="num-reversed">14</span></span>
             <span>of 191 alphas</span>
           </div>
           <div class="stat dead">
             <span class="label">Dead</span>
-            <span class="num"><span class="num-dead">165</span></span>
+            <span class="num"><span class="num-dead">168</span></span>
             <span>of 191 alphas</span>
           </div>
         </div>
-        <p class="callout">Numbers below are the live W4.a bench output on the bundled gtja191 zoo (CSI 300, 2018&ndash;2025) &mdash; reproducible via the CLI snippet at the end of the post.</p>
+        <p class="callout">Numbers below are from a bench run on 2026-09-29 (Vibe-Trading 0.1.16: CSI 300 with point-in-time membership and split-adjusted prices, 2018&ndash;2025) &mdash; reproducible via the CLI snippet at the end of the post.</p>
       </section>
 
       <section id="background">
@@ -294,8 +297,8 @@
         <p>The 2014 Guotai Junan research report &mdash; titled "191 个短周期交易型 alpha 因子"
         (<em>191 Short-Period Transactional Alpha Factors</em>) &mdash; landed at a particular moment in
         Chinese quant. The market had just emerged from a multi-year sideways grind, retail participation
-        was climbing back, and the Kakushadze-style "formulaic alpha" template had been popularised
-        internationally a few years earlier by his 2015 arXiv preprint. The GTJA team produced what is,
+        was climbing back, and the "formulaic alpha" template would be popularised internationally a
+        year later by Kakushadze's 2015 arXiv preprint, <em>101 Formulaic Alphas</em>. The GTJA team produced what is,
         in retrospect, one of the most systematic public catalogues of short-horizon factors ever published
         for the A-share universe: 191 numbered formulas, each a few lines of operator algebra over daily
         OHLCV plus turnover.</p>
@@ -313,7 +316,7 @@
         changed: T+1 settlement remains, but the institutional/retail mix has flipped, the STAR Board and
         ChiNext registration-system reform have re-priced small-cap risk, and high-frequency-flavoured
         execution by mutual funds has compressed many obvious mean-reversion windows.</p>
-        <p>So we ran the test. The methodology preview is below; the numbers will follow.</p>
+        <p>So we ran the test: method first, then the numbers.</p>
       </section>
 
       <section id="method">
@@ -325,11 +328,10 @@
         question and is out of scope for this post.</p>
 
         <h3>Universe</h3>
-        <p>CSI 300. The 300 most liquid A-shares, rebalanced semi-annually by the index provider. We use
-        the current index constituents on each rebalance date, so the universe drifts with the index. This
-        is biased relative to a frozen universe (the dropouts from CSI 300 are precisely the names whose
-        liquidity has decayed, which can bias the IC estimate), but it is consistent with how an actual
-        deployment would use the index.</p>
+        <p>CSI 300: the 300 most liquid A-shares, rebalanced semi-annually by the index provider. Each
+        day's cross-section is the index membership in force on that day, from Tushare's month-end
+        rosters, so a stock counts only while it is in the index &mdash; 348 names were members at some
+        point in the window. Prices are forward-adjusted for splits and dividends.</p>
 
         <h3>Period</h3>
         <p>2018-01-02 through 2025-12-31. Eight calendar years, ~1,940 trading days, fully out-of-sample
@@ -340,7 +342,7 @@
         <h3>Signal definition</h3>
         <p>For each alpha and each trading day <em>t</em>, we compute the alpha value for every stock in
         the universe, then cross-sectionally rank-transform it to <code class="formula">[0, 1]</code>.
-        Forward return is the 1-day log return from close <em>t</em> to close <em>t+1</em>, also
+        Forward return is the 1-day simple return from close <em>t</em> to close <em>t+1</em>, also
         cross-sectionally rank-transformed. The Spearman IC for day <em>t</em> is the Pearson correlation
         of the two ranked series. We report the mean IC across all valid days, the t-statistic of the IC
         series, and the fraction of days with positive IC.</p>
@@ -352,8 +354,9 @@
           <li><strong>Reversed.</strong> Mean IC &lt; &minus;0.02 and t-stat(IC) &lt; &minus;2 (sign-flipped
           version of the above). The original report intended the alpha to predict with one sign; we
           observe it predicting with the opposite sign at statistical significance.</li>
-          <li><strong>Dead.</strong> Everything else: |mean IC| &le; 0.02, or |t-stat| &le; 2, or the alpha
-          could not be computed cleanly over the window (e.g. requires intraday tick data we do not have).</li>
+          <li><strong>Dead.</strong> Everything else that could be computed: |mean IC| &le; 0.02, or
+          |t-stat| &le; 2, or a positive IC on fewer than 55% of days. An alpha whose output is more than
+          95% NaN is not tested at all and is reported separately.</li>
         </ul>
 
         <h3>Caveats baked into the method</h3>
@@ -363,54 +366,53 @@
           report's preferred horizon) will produce slightly different numbers; we'll publish those as a
           follow-up.</li>
           <li>No t-cost adjustment. A high-IC alpha with daily-rebalance turnover near 100% is, in
-          practice, unprofitable after a realistic 5&ndash;10 bps round-trip. We report turnover alongside
-          IC in the artefact bench output, but the survival classification here ignores it.</li>
-          <li>No sector neutralisation. Some alphas in the report rely on sector mapping
-          (<code class="formula">indneutralize</code>); for those we substitute the per-day
-          cross-sectional mean as a fallback, which is a strictly weaker neutralisation. Affected alphas
-          are flagged in the per-alpha output.</li>
+          practice, unprofitable after a realistic 5&ndash;10 bps round-trip. The bench does not
+          measure turnover, and the survival classification here ignores it.</li>
+          <li>No sector neutralisation. None of the GTJA 191 formulas needs sector tags, so every IC here
+          is measured on the raw cross-section.</li>
         </ul>
 
         <h3>Tool used</h3>
         <pre><code>$ vibe-trading alpha bench --zoo gtja191 --universe csi300 \
                             --period 2018-2025 --top 20</code></pre>
-        <p>One command. Output is an HTML report with per-alpha IC, t-stat, turnover, and IC time-series
-        plot. Reproducibility recipe at the end of the post.</p>
+        <p>One command. It writes an HTML report to <code>~/.vibe-trading/reports/</code> with the top
+        alphas by IR (mean IC, IC standard deviation, IR, share of positive-IC days, number of days) and
+        the alphas it could not compute. Reproducibility recipe at the end of the post.</p>
       </section>
 
       <section id="findings">
         <h2>Findings</h2>
 
         <h3>Aggregate survival</h3>
-        <p>The headline counts again, with the same caveat that final numbers land after the W4.a bench:</p>
+        <p>The headline counts again:</p>
         <div class="findings-stats" aria-label="Aggregate alpha survival counts, repeated for findings section">
           <div class="stat alive">
             <span class="label">Alive</span>
-            <span class="num"><span class="num-alive">10</span></span>
+            <span class="num"><span class="num-alive">8</span></span>
             <span>survive all three filters</span>
           </div>
           <div class="stat reversed">
             <span class="label">Reversed</span>
-            <span class="num"><span class="num-reversed">15</span></span>
+            <span class="num"><span class="num-reversed">14</span></span>
             <span>now predict with opposite sign</span>
           </div>
           <div class="stat dead">
             <span class="label">Dead</span>
-            <span class="num"><span class="num-dead">165</span></span>
+            <span class="num"><span class="num-dead">168</span></span>
             <span>below significance threshold</span>
           </div>
         </div>
-        <p>What surprised us, in early partial runs, is not the dead count &mdash; everyone expects
-        decay &mdash; but the <em>reversed</em> count. A meaningful slice of alphas that worked in 2014
-        now act as contrarian indicators with statistically significant magnitude. The simplest reading is
-        that a behavioural anomaly the formula was capturing (small-cap mean reversion, end-of-day flow
-        from retail) has been crowded out by precisely the kind of systematic trading the formula
-        represents, and what remains is the opposite trade.</p>
+        <p>What surprised us is not the dead count &mdash; everyone expects decay &mdash; but the
+        <em>reversed</em> count. A meaningful slice of alphas that worked in 2014 now act as contrarian
+        indicators with statistically significant magnitude. The simplest reading is that a behavioural
+        anomaly the formula was capturing (small-cap mean reversion, end-of-day flow from retail) has been
+        crowded out by precisely the kind of systematic trading the formula represents, and what remains
+        is the opposite trade.</p>
 
         <h3>Theme breakdown</h3>
-        <p>We hand-tagged each alpha by its dominant theme &mdash; the operator vocabulary it leans on
-        most heavily. This is a lossy categorisation (many alphas blend two themes), but the aggregate
-        pattern is robust to the labelling choice. Numbers below are survival rates within each theme:</p>
+        <p>Each alpha carries one or two theme tags in its registry metadata &mdash; the operator
+        vocabulary it leans on most heavily. An alpha with two tags counts in both rows, so the counts add
+        up to more than 190. Numbers below are survival rates within each theme:</p>
 
         <table class="theme-table" aria-label="Theme-by-theme survival rate">
           <thead>
@@ -426,7 +428,7 @@
               <td>Volume-price interaction</td>
               <td>Correlation / covariance of volume with close, high-low range</td>
               <td>81</td>
-              <td>5% (4/81)</td>
+              <td>4% (3/81)</td>
             </tr>
             <tr>
               <td>Short-horizon volatility</td>
@@ -438,13 +440,13 @@
               <td>Reversal</td>
               <td>Negative-sign return signals over 1-5 day horizons</td>
               <td>38</td>
-              <td>11% (4/38)</td>
+              <td>8% (3/38)</td>
             </tr>
             <tr>
               <td>Momentum</td>
               <td>Positive-sign return signals over 10-60 day horizons</td>
               <td>63</td>
-              <td>2% (1/63)</td>
+              <td>0% (0/63)</td>
             </tr>
             <tr>
               <td>Turnover / liquidity</td>
@@ -458,79 +460,84 @@
               <td>18</td>
               <td>22% (4/18)</td>
             </tr>
+            <tr>
+              <td>Sentiment</td>
+              <td>Share of up days on days the market fell (one alpha, <code>gtja191_075</code>, on a proxy benchmark)</td>
+              <td>1</td>
+              <td>0% (0/1)</td>
+            </tr>
           </tbody>
         </table>
 
-        <p>The actual read from the W4.a bench: <em>microstructure / range</em> alphas are the standout
-        survivors (22% survival, 4 of 18), with <em>reversal</em> next (11%, 4 of 38). The categories
-        we expected to age best &mdash; raw <em>volume-price interaction</em> (5%) and
-        <em>short-horizon volatility</em> (8%) &mdash; have decayed more than the open-close-high-low
-        decomposition family. <em>Momentum</em> at 2% (1 of 63) and the small <em>turnover</em> bucket
-        at 0% are the cleanly-arbitraged groups. The pattern is consistent with what you would expect
+        <p>The read from the re-run: <em>microstructure / range</em> alphas are the standout survivors
+        (22%, 4 of 18). <em>Reversal</em> and <em>short-horizon volatility</em> follow at 8% (3 of 38 and
+        2 of 26), raw <em>volume-price interaction</em> at 4% (3 of 81). <em>Momentum</em> keeps none of
+        its 63 &mdash; five of them now predict with the opposite sign &mdash; and the small
+        <em>turnover</em> bucket keeps none of its 2. The pattern is consistent with what you would expect
         if a decade of systematic capital has compressed the easiest reversal/momentum trades but left
         intact the structural daily-bar geometry alphas that key on shape-of-the-bar effects.</p>
 
         <h3>Top 5 surviving alphas</h3>
-        <p>Five alphas with the highest mean IC across the window, in descending order. Each is identified
-        by its zoo id and a paraphrase of its formula; full operator-level definition is in the
-        <code>__alpha_meta__["formula_latex"]</code> field of the corresponding Python module.</p>
+        <p>The five alphas with the highest IR (mean IC divided by its standard deviation) across the
+        window, in descending order, each with its formula as the registry stores it
+        (<code>__alpha_meta__["formula_latex"]</code> of the corresponding Python module).</p>
 
-                <article class="alpha-card">
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_114</span>
+          <pre class="formula-block"><code>((RANK(DELAY(((HIGH-LOW)/(SUM(CLOSE,5)/5)),2)) * RANK(RANK(VOLUME))) / (((HIGH-LOW)/(SUM(CLOSE,5)/5))/(VWAP-CLOSE)))</code></pre>
+          <p class="alpha-note">Mean IC = 0.0606, IR = 0.2926 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_114</code>).</p>
+        </article>
+        <article class="alpha-card">
           <span class="alpha-id">gtja191_171</span>
           <pre class="formula-block"><code>-1*((l-c)*(o^5))/((c-h)*(c^5))</code></pre>
-          <p class="alpha-note">Mean IC = 0.0432, IR = 0.2690 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_171</code>).</p>
+          <p class="alpha-note">Mean IC = 0.0435, IR = 0.2299 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_171</code>).</p>
+        </article>
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_120</span>
+          <pre class="formula-block"><code>rank(vwap-close)/rank(vwap+close)</code></pre>
+          <p class="alpha-note">Mean IC = 0.0452, IR = 0.2119 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_120</code>).</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_111</span>
           <pre class="formula-block"><code>sma(v*((c-l)-(h-c))/(h-l),11,2)-sma(v*((c-l)-(h-c))/(h-l),4,2)</code></pre>
-          <p class="alpha-note">Mean IC = 0.0349, IR = 0.2232 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_111</code>).</p>
+          <p class="alpha-note">Mean IC = 0.0355, IR = 0.1888 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_111</code>).</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_163</span>
           <pre class="formula-block"><code>rank(((-1*ret)*mean(v,20))*vwap*(high-close))</code></pre>
-          <p class="alpha-note">Mean IC = 0.0347, IR = 0.2008 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_163</code>).</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_002</span>
-          <pre class="formula-block"><code>(-1 * DELTA(((CLOSE - LOW) - (HIGH - CLOSE)) / (HIGH - LOW), 1))</code></pre>
-          <p class="alpha-note">Mean IC = 0.0262, IR = 0.1619 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_002</code>).</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_054</span>
-          <pre class="formula-block"><code>((-1*RANK((STD(ABS(CLOSE-OPEN),10)+(CLOSE-OPEN))+CORR(CLOSE,OPEN,10))))</code></pre>
-          <p class="alpha-note">Mean IC = 0.0272, IR = 0.1606 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_054</code>).</p>
+          <p class="alpha-note">Mean IC = 0.0342, IR = 0.1674 over the CSI 300 / 2018&ndash;2025 window. Formula reproduced verbatim from the registry (<code>__alpha_meta__["formula_latex"]</code> of <code>gtja191_163</code>).</p>
         </article>
 
-        <h3>Three famously dead alphas</h3>
-        <p>Three formulas that worked, or were claimed to work, in the 2014 report but now sit comfortably
-        in the dead or reversed bucket:</p>
+        <h3>Three alphas that now point the other way</h3>
+        <p>The three alphas with the most negative mean IC. All three sit in the reversed bucket: they now
+        predict, with statistical significance, in the direction opposite to the report's sign.</p>
 
-                <article class="alpha-card">
+        <article class="alpha-card">
+          <span class="alpha-id">gtja191_124 (reversed)</span>
+          <pre class="formula-block"><code>(close-vwap)/decay_linear(rank(tsmax(close,30)),2)</code></pre>
+          <p class="alpha-note">Mean IC = -0.0518, IR = -0.2516. The most negative mean IC in the gtja191 zoo on CSI 300 / 2018&ndash;2025.</p>
+        </article>
+        <article class="alpha-card">
           <span class="alpha-id">gtja191_178 (reversed)</span>
           <pre class="formula-block"><code>(c-delay(c,1))/delay(c,1)*v</code></pre>
-          <p class="alpha-note">Mean IC = -0.0327, IR = -0.1930. Worst-performing slice of the gtja191 zoo on CSI 300 / 2018&ndash;2025 by raw IC.</p>
+          <p class="alpha-note">Mean IC = -0.0329, IR = -0.1644. The second most negative.</p>
         </article>
         <article class="alpha-card">
           <span class="alpha-id">gtja191_137 (reversed)</span>
-          <pre class="formula-block"><code>see body</code></pre>
-          <p class="alpha-note">Mean IC = -0.0277, IR = -0.1556. Worst-performing slice of the gtja191 zoo on CSI 300 / 2018&ndash;2025 by raw IC.</p>
-        </article>
-        <article class="alpha-card">
-          <span class="alpha-id">gtja191_066 (reversed)</span>
-          <pre class="formula-block"><code>(CLOSE-MEAN(CLOSE,6))/MEAN(CLOSE,6)*100</code></pre>
-          <p class="alpha-note">Mean IC = -0.0270, IR = -0.1377. Worst-performing slice of the gtja191 zoo on CSI 300 / 2018&ndash;2025 by raw IC.</p>
+          <pre class="formula-block"><code>16*((c-dc1+(c-o)/2+dc1-do1)/MAX_term) * MAX(abs(h-dc1), abs(l-dc1))</code></pre>
+          <p class="alpha-note">Mean IC = -0.0300, IR = -0.1464. The third most negative. <code>MAX_term</code> is the report's three-way piecewise denominator, built from |h &minus; dc1|, |l &minus; dc1|, |h &minus; dl1| and |dc1 &minus; do1| (dc1, do1, dl1: the previous day's close, open and low); the module carries the full definition.</p>
         </article>
 
         <h3>One paragraph of reflection</h3>
         <p>The temptation, after a survival study like this, is to over-generalise: "decay is inevitable;
         formulaic alphas are dead." We don't think that's the right read. What this exercise teaches is
-        narrower and more useful: a meaningful fraction of a 12-year-old short-horizon catalogue still
-        produces signal, the survivors cluster in interpretable themes (volume-price interaction,
-        short-horizon volatility), and the dead ones cluster in equally interpretable themes (naive
-        reversal, simple turnover transforms). It tells us very little about whether the Kakushadze 101
-        formulas, or the Qlib 158 feature set, will decay at the same rate &mdash; those zoos have
-        different operator vocabularies and different intended universes (US equities, multi-horizon
-        respectively). Future work, separate bench runs.</p>
+        narrower and more useful: a small part of a 12-year-old short-horizon catalogue &mdash; 8 of 190
+        &mdash; still produces signal; the survivors cluster in an interpretable theme (bar geometry:
+        open-close-high-low decompositions), and the losers cluster in equally interpretable ones
+        (momentum, where nothing survives and five alphas flipped sign, and simple turnover transforms).
+        It tells us very little about whether the Kakushadze 101 formulas, or the Qlib 158 feature set,
+        will decay at the same rate &mdash; those zoos have different operator vocabularies and different
+        intended universes (US equities, multi-horizon respectively). Future work, separate bench runs.</p>
       </section>
 
       <section id="caveats">
@@ -554,11 +561,13 @@
         muted), and some "dead" alphas here might revive on a CSI 1000 cut.</p>
 
         <h3>Tushare data scope</h3>
-        <p>Our data feed (Tushare end-of-day OHLCV plus turnover, plus VWAP derived from amount) does not
-        include intraday tick or order-book information. A small number of GTJA 191 alphas in their
-        original formulation reference intraday quantities (depth-weighted VWAP, level-1 book imbalance);
-        for those we substitute the daily VWAP and flag the substitution in the per-alpha output. Those
-        alphas should be considered <em>not properly tested</em> here, not "dead".</p>
+        <p>Our data feed is Tushare end-of-day OHLCV and traded amount, forward-adjusted for splits and
+        dividends, with VWAP derived from amount; it has no intraday tick or order-book data. Five GTJA 191
+        formulas reference an input the panel does not carry &mdash; the benchmark index close
+        (<code>gtja191_075</code>, <code>_149</code>, <code>_181</code>, <code>_182</code>) or a
+        multi-factor regression residual (<code>gtja191_030</code>) &mdash; and use a proxy instead: the
+        cross-sectional mean close, or a weighted average of squared daily returns, as each alpha's notes
+        say. Treat those five as <em>not properly tested</em> here, not as "dead".</p>
 
         <h3>8-year window is short</h3>
         <p>By academic standards an 8-year window is on the short side. The classic Fama-French papers
@@ -568,16 +577,21 @@
         window", not "dead forever".</p>
 
         <h3>Survivorship bias in the universe</h3>
-        <p>The CSI 300 constituent list used here is the <em>current</em> index membership applied across
-        the full 2018&ndash;2025 window, not a point-in-time reconstruction of the index on each rebalance.
-        Stocks that were delisted, demoted, or removed during the window are absent; stocks added late get
-        their full price history backfilled. This is a standard form of survivorship bias and it
-        biases IC estimates upward in the long-only direction. The same caveat applies even more strongly
-        to any cross-zoo comparison against a US universe constructed the same way &mdash; for example, a
-        SP 500 run using today's constituent list will overstate alive counts relative to a true
-        point-in-time universe, and a 0% alive count there should be read as "decay plus survivorship
-        cancelled out the signal", not as a clean failure of the alpha family. Point-in-time index
-        membership is a planned upgrade.</p>
+        <p>The first version of this post applied the latest CSI 300 roster to the whole 2018&ndash;2025
+        window, so every name was present because it survived to the end &mdash; a hindsight-selected set,
+        which biases IC upward. The re-run masks each day's cross-section to the roster in force that day
+        (348 names were members at some point in the window). The <code>sp500</code> universe still uses
+        today's constituent list, so the same caveat applies to any cross-zoo comparison run on it &mdash;
+        its bench report says so &mdash; and a low alive count there should be read as "decay plus
+        survivorship", not as a clean failure of the alpha family.</p>
+
+        <h3>190 formulas, one test each</h3>
+        <p>Testing 190 formulas on the same window makes some look good by chance. The bench reports how
+        large the best IR would be expected to be if none of them carried any signal: 0.204 here. The best
+        observed IR, 0.293 (<code>gtja191_114</code>), survives that deflation (probability 0.9999). Two
+        more alive alphas are above 0.204 &mdash; <code>gtja191_171</code> (0.230) and
+        <code>gtja191_120</code> (0.212); the other five, with IRs between 0.130 and 0.189, are below it,
+        so read them as candidates rather than confirmed signals.</p>
 
         <h3>No regime conditioning</h3>
         <p>We average IC over the full 2018&ndash;2025 window. Many of these alphas are almost certainly
@@ -597,11 +611,11 @@
 export TUSHARE_TOKEN=your_token_here
 vibe-trading alpha bench --zoo gtja191 \
                          --universe csi300 \
-                         --period 2020-2025 \
+                         --period 2018-2025 \
                          --top 20</code></pre>
 
-        <p>You get back an HTML report saved to <code>~/.vibe-trading/reports/</code> with per-alpha IC,
-        IR, decay curve, and a sortable top-N table. The same command works against
+        <p>You get back an HTML report saved to <code>~/.vibe-trading/reports/</code> with the top alphas
+        by IR and the alphas that could not be computed. The same command works against
         <code>--zoo alpha101</code> (Kakushadze 101 Formulaic Alphas, paper-faithful rewrite of the 2015
         arXiv preprint) and <code>--zoo qlib158</code> (the Microsoft Qlib feature library, used under
         Apache-2.0 with attribution). Cross-zoo comparison runs via <code>alpha compare</code>.</p>
@@ -613,7 +627,7 @@ vibe-trading alpha bench --zoo gtja191 \
         </div>
 
         <p>If you find an alpha whose survival classification surprises you &mdash; especially a survivor
-        we did not flag in the top-5 or a famously-dead one we kept in the alive bucket &mdash; please
+        we did not flag in the top 5, or a reversed one you expected to survive &mdash; please
         open an issue with the alpha id and your reasoning. Community pull requests adding new zoos, new
         universes (CSI 1000, NASDAQ 100, crypto majors) or new validation tooling are welcome under the
         <a href="https://github.com/HKUDS/Vibe-Trading/blob/main/CONTRIBUTING.md">CONTRIBUTING.md</a>
diff --git a/wiki/scripts/build_alpha_library.py b/wiki/scripts/build_alpha_library.py
index 7b192384..28699776 100644
--- a/wiki/scripts/build_alpha_library.py
+++ b/wiki/scripts/build_alpha_library.py
@@ -23,6 +23,9 @@ Security posture (v0.1.8):
     * KaTeX is intentionally **not** loaded. ``formula_latex`` is shown as
       raw escaped text inside ``<pre>``. A future revision can swap in a
       server-side LaTeX→MathML pass that keeps CSP intact.
+    * Notes, nicknames and theme tags carry their Chinese in ``data-text-zh``
+      from ``wiki/alpha-library/i18n.zh.json`` — only when the stored English
+      still matches the manifest, so a stale translation is never shown.
     * The header, footer and head block come from ``wiki/partials/`` (the
       same markup ``sync_site_chrome.py`` writes into the hand-written
       pages); labels carry ``data-i18n`` keys from ``wiki/locales/zh.json``.
@@ -68,6 +71,11 @@ except ImportError as exc:  # pragma: no cover — Jinja2 is a project dep
 _REPO_ROOT = Path(__file__).resolve().parents[2]
 _DEFAULT_MANIFEST = _REPO_ROOT / "wiki" / "alpha-library" / "manifest.json"
 _DEFAULT_OUTPUT = _REPO_ROOT / "wiki" / "alpha-library" / "content"
+# Chinese notes, nicknames and theme names. Each note and nickname keeps the
+# English it was translated from, so an edited English text is shown in
+# English until its translation is updated (agent/tests/factors/
+# test_alpha_library_translations.py fails in the meantime).
+_I18N_PATH = _REPO_ROOT / "wiki" / "alpha-library" / "i18n.zh.json"
 
 # Strict CSP for static content pages: same-origin scripts only, no inline.
 _CSP_CONTENT = (
@@ -80,7 +88,7 @@ _CSP_CONTENT = (
 )
 
 # Cache-busting version of the site shell modules; matches the hand-written pages.
-_SHELL_VERSION = "20260929d"
+_SHELL_VERSION = "20260929e"
 
 # Display names for zoos. Source of truth — do not depend on prose elsewhere.
 _ZOO_DISPLAY: dict[str, dict[str, str]] = {
@@ -199,11 +207,11 @@ _ALPHA_PAGE_TEMPLATE = """<!doctype html>
       <span>{{ alpha.id }}</span>
     </p>
     <h1>{{ alpha.id }}</h1>
-    {% if alpha.meta.nickname %}<p class="nickname">{{ alpha.meta.nickname }}</p>{% endif %}
+    {% if alpha.meta.nickname %}<p class="nickname"{% if zh.nickname %} data-text-zh="{{ zh.nickname }}"{% endif %}>{{ alpha.meta.nickname }}</p>{% endif %}
 
     <h2 data-i18n="alpha.page.themes">Themes &amp; universe</h2>
     <p>
-      {% for t in alpha.meta.theme %}<span class="tag">{{ t }}</span>{% endfor %}
+      {% for t in alpha.meta.theme %}<span class="tag" data-text-zh="{{ t | zh_theme }}">{{ t }}</span>{% endfor %}
       {% for u in alpha.meta.universe %}<span class="tag">{{ u }}</span>{% endfor %}
       {% for f in alpha.meta.frequency %}<span class="tag">{{ f }}</span>{% endfor %}
     </p>
@@ -225,7 +233,7 @@ _ALPHA_PAGE_TEMPLATE = """<!doctype html>
 
     {% if alpha.meta.notes %}
     <h2 data-i18n="alpha.page.notes">Notes</h2>
-    <p class="notes">{{ alpha.meta.notes }}</p>
+    <p class="notes"{% if zh.notes %} data-text-zh="{{ zh.notes }}"{% endif %}>{{ alpha.meta.notes }}</p>
     {% endif %}
 
     <h2 data-i18n="alpha.page.run">Run it</h2>
@@ -290,7 +298,7 @@ _ZOO_PAGE_TEMPLATE = """<!doctype html>
       {% for a in zoo.alphas %}
         <tr>
           <td><a href="{{ a.id }}.html">{{ a.id }}</a></td>
-          <td class="theme-cell">{{ a.meta.theme | join(", ") }}</td>
+          <td class="theme-cell" data-text-zh="{{ a.meta.theme | map('zh_theme') | join('、') }}">{{ a.meta.theme | join(", ") }}</td>
           <td class="theme-cell">{{ a.meta.universe | join(", ") }}</td>
           <td>{{ a.meta.decay_horizon }}</td>
           <td>{{ a.meta.min_warmup_bars }}</td>
@@ -311,13 +319,47 @@ _ZOO_PAGE_TEMPLATE = """<!doctype html>
 # ---------------------------------------------------------------------------
 
 
+def _load_i18n() -> dict[str, Any]:
+    """Load the Chinese notes / nicknames / theme names (empty when absent).
+
+    Returns:
+        ``{"notes": {id: {"en", "zh"}}, "nicknames": {...}, "themes": {tag: zh}}``.
+    """
+    if not _I18N_PATH.is_file():
+        return {"notes": {}, "nicknames": {}, "themes": {}}
+    return json.loads(_I18N_PATH.read_text(encoding="utf-8"))
+
+
+_I18N = _load_i18n()
+
+
+def _zh_texts(alpha: dict) -> dict[str, str | None]:
+    """Chinese note and nickname for one alpha, or None where stale / missing.
+
+    Args:
+        alpha: One manifest alpha entry.
+
+    Returns:
+        ``{"notes": ..., "nickname": ...}``.
+    """
+    meta = alpha.get("meta", {})
+    out: dict[str, str | None] = {}
+    for field, bucket in (("notes", "notes"), ("nickname", "nicknames")):
+        entry = _I18N.get(bucket, {}).get(alpha.get("id"))
+        text = meta.get(field)
+        out[field] = entry["zh"] if entry and text and entry.get("en") == text else None
+    return out
+
+
 def _build_env() -> Environment:
     """Create a Jinja2 environment with mandatory autoescape on."""
-    return Environment(
+    env = Environment(
         autoescape=select_autoescape(["html", "xml"]),
         trim_blocks=True,
         lstrip_blocks=True,
     )
+    env.filters["zh_theme"] = lambda tag: _I18N.get("themes", {}).get(tag, tag)
+    return env
 
 
 def _chrome() -> dict[str, Any]:
@@ -351,6 +393,7 @@ def render_alpha_page(env: Environment, alpha: dict, zoo: dict) -> str:
         zoo=zoo,
         zoo_display=zoo_display,
         csp=_CSP_CONTENT,
+        zh=_zh_texts(alpha),
         **_chrome(),
     )
 
@@ -472,6 +515,7 @@ def main(argv: list[str] | None = None) -> int:
     zoo_count = 0
     sample_alpha: dict | None = None
     sample_html: str | None = None
+    untranslated: list[str] = []
 
     for zoo in manifest.get("zoos", []):
         zoo_id = zoo.get("zoo_id")
@@ -492,6 +536,10 @@ def main(argv: list[str] | None = None) -> int:
                 # regex, but never trust a JSON file from disk.
                 continue
             html = render_alpha_page(env, alpha, zoo)
+            zh = _zh_texts(alpha)
+            meta = alpha.get("meta", {})
+            if (meta.get("notes") and not zh["notes"]) or (meta.get("nickname") and not zh["nickname"]):
+                untranslated.append(alpha_id)
             _write(zoo_dir / f"{alpha_id}.html", html)
             alpha_count += 1
             if sample_alpha is None:
@@ -513,6 +561,12 @@ def main(argv: list[str] | None = None) -> int:
             print("html escape FAIL", file=sys.stderr)
             return 1
 
+    if untranslated:
+        print(
+            f"warning: {len(untranslated)} alphas show English notes/nicknames in Chinese "
+            f"(missing or stale in {_I18N_PATH.name}): {', '.join(untranslated[:10])}",
+            file=sys.stderr,
+        )
     print(
         f"built: {alpha_count} alphas across {zoo_count} zoos -> {out_dir}"
     )
diff --git a/wiki/site.js b/wiki/site.js
index 74a9e570..18f0fe94 100644
--- a/wiki/site.js
+++ b/wiki/site.js
@@ -4,8 +4,9 @@
 // Language model: English is the HTML itself. Chinese comes from
 // /locales/zh.json through data-i18n (text), data-i18n-html (markup),
 // data-i18n-attr ("attr=key,attr=key") and data-href-zh (a link that has a
-// Chinese counterpart). One stored choice, default English, never guessed
-// from the browser.
+// Chinese counterpart); generated pages carry their own Chinese inline in
+// data-text-zh (the Alpha Library's per-alpha notes). One stored choice,
+// default English, never guessed from the browser.
 //
 // A page whose URL is language-neutral (<html data-lang-neutral>) renders the
 // stored language. A page whose URL is language-specific (<html
@@ -14,7 +15,7 @@
 // site Chinese; its toggle goes to the counterpart page instead.
 
 const LANG_KEY = "vibetrading-lang";
-const LOCALE_VERSION = "20260929d";
+const LOCALE_VERSION = "20260929e";
 const THEME_KEY = "vibetrading-theme";
 const STARS_CACHE_KEY = "vibetrading-github-stars";
 const STARS_TTL_MS = 12 * 60 * 60 * 1000;
@@ -79,6 +80,11 @@ function translateElement(el, messages) {
       el.setAttribute(attr, messages?.[key] ?? original);
     }
   }
+  const textZh = el.getAttribute("data-text-zh");
+  if (textZh) {
+    const original = remember(el, "i18nEnText", el.textContent);
+    el.textContent = messages ? textZh : original;
+  }
   const hrefZh = el.getAttribute("data-href-zh");
   if (hrefZh) {
     const original = remember(el, "i18nEnHref", el.getAttribute("href") ?? "");
@@ -98,7 +104,7 @@ export async function translate(lang, root = document) {
       messages = null;
     }
   }
-  root.querySelectorAll("[data-i18n], [data-i18n-html], [data-i18n-attr], [data-href-zh]").forEach((el) => {
+  root.querySelectorAll("[data-i18n], [data-i18n-html], [data-i18n-attr], [data-href-zh], [data-text-zh]").forEach((el) => {
     translateElement(el, messages);
   });
   if (root === document) {
diff --git a/wiki/tutorials/index.html b/wiki/tutorials/index.html
index d55167ca..fc789b97 100644
--- a/wiki/tutorials/index.html
+++ b/wiki/tutorials/index.html
@@ -18,10 +18,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
 </head>
 <body data-section="tutorials">
   <!-- site-header:start -->
diff --git a/wiki/tutorials/vibe-trading-beginner-zh.html b/wiki/tutorials/vibe-trading-beginner-zh.html
index a7992e6d..e1938a67 100644
--- a/wiki/tutorials/vibe-trading-beginner-zh.html
+++ b/wiki/tutorials/vibe-trading-beginner-zh.html
@@ -21,10 +21,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
   <style>
     .tutorial {
       width: min(940px, calc(100% - 32px));
diff --git a/wiki/tutorials/vibe-trading-beginner.html b/wiki/tutorials/vibe-trading-beginner.html
index 1562e1a5..91d0839f 100644
--- a/wiki/tutorials/vibe-trading-beginner.html
+++ b/wiki/tutorials/vibe-trading-beginner.html
@@ -21,10 +21,10 @@
   <link rel="preconnect" href="https://fonts.googleapis.com">
   <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
   <link href="https://fonts.googleapis.com/css2?family=Fraunces:opsz,wght@9..144,600;9..144,750&amp;family=IBM+Plex+Mono:wght@400;500;600&amp;family=IBM+Plex+Sans:wght@400;500;600;700&amp;display=swap" rel="stylesheet">
-  <link rel="stylesheet" href="/styles.css?v=20260929d">
-  <script src="/theme-init.js?v=20260929d"></script>
+  <link rel="stylesheet" href="/styles.css?v=20260929e">
+  <script src="/theme-init.js?v=20260929e"></script>
   <!-- site-head:end -->
-  <script type="module" src="/main.js?v=20260929d"></script>
+  <script type="module" src="/main.js?v=20260929e"></script>
   <style>
     .tutorial {
       width: min(940px, calc(100% - 32px));