diff --git a/research/arxiv_parsed.txt b/research/arxiv_parsed.txt
deleted file mode 100644
index e69de29..0000000
diff --git a/research/arxiv_results.xml b/research/arxiv_results.xml
deleted file mode 100644
index e69de29..0000000
diff --git a/research/data/README.md b/research/data/README.md
new file mode 100644
index 0000000..b2ff162
--- /dev/null
+++ b/research/data/README.md
@@ -0,0 +1,18 @@
+# Research Data
+
+本目录存放论文调研的原始数据文件和缓存。
+
+## 文件说明
+
+| 文件 | 说明 |
+|------|------|
+| `search_results.json` | arXiv + GitHub 历史检索缓存（ZED2i 相关，含 2026-04/05 最新论文） |
+| `crowdroom_papers_raw.json` | CrowdRoom 专属 arXiv 抓取结果（当前因 429 限速为空） |
+| `lyra2_paper.txt` | Lyra 2.0 论文全文（arXiv:2604.13036，NVIDIA，2026-04） |
+| `lyra2_paper.pdf` | Lyra 2.0 论文 PDF 原文 |
+
+## 注意事项
+
+- `search_results.json` 是重要的历史缓存，请勿随意删除
+- 数据文件不纳入 git 版本控制（见 `.gitignore`）
+- 重新抓取请使用 `../tools/` 目录下的脚本
diff --git a/research/crowdroom_papers_raw.json b/research/data/crowdroom_papers_raw.json
similarity index 100%
rename from research/crowdroom_papers_raw.json
rename to research/data/crowdroom_papers_raw.json
diff --git a/research/data/lyra2_paper.pdf b/research/data/lyra2_paper.pdf
new file mode 100644
index 0000000..41b4aa3
Binary files /dev/null and b/research/data/lyra2_paper.pdf differ
diff --git a/research/lyra2_paper.txt b/research/data/lyra2_paper.txt
similarity index 100%
rename from research/lyra2_paper.txt
rename to research/data/lyra2_paper.txt
diff --git a/research/search_results.json b/research/data/search_results.json
similarity index 100%
rename from research/search_results.json
rename to research/data/search_results.json
diff --git a/research/index.md b/research/index.md
index 84753eb..6d115ec 100644
--- a/research/index.md
+++ b/research/index.md
@@ -11,7 +11,7 @@ description: "研究论文调研综述目录，涵盖世界模型、空间感知
 
 本索引汇总了 **WorldModel 项目**下所有论文调研综述文档。项目围绕「可探索生成式 3D 世界」、「空间感知与记忆」、「物理世界理解」、「传感器数据管道」以及「具体子项目（CrowdRoom / PRISM）」五大方向展开研究，旨在为 PRISM 2.0、CrowdRoom MVP 等工程系统提供学术文献支撑与技术选型依据。
 
-> **更新时间**：2026-05-20 | **文档数量**：9 篇 | **覆盖论文**：~215 篇
+> **更新时间**：2026-05-20 | **文档数量**：8 篇 | **覆盖论文**：~215 篇
 
 ---
 
@@ -92,6 +92,32 @@ description: "研究论文调研综述目录，涵盖世界模型、空间感知
 
 ---
 
+## 工具与数据
+
+### 脚本工具（`tools/`）
+
+| 脚本 | 用途 |
+|------|------|
+| [`tools/search_info.py`](tools/search_info.py) | arXiv + GitHub 主题化检索，输出 `data/search_results.json` |
+| [`tools/fetch_crowdroom_papers.py`](tools/fetch_crowdroom_papers.py) | CrowdRoom 专属 arXiv 抓取，输出 `data/crowdroom_papers_raw.json` |
+| [`tools/gen_review_from_json.py`](tools/gen_review_from_json.py) | 将检索缓存渲染为 ZED2i 综述 Markdown |
+| [`tools/_build_crowdroom_review.py`](tools/_build_crowdroom_review.py) | 将抓取数据构建为 CrowdRoom 综述 Markdown |
+
+详见 [`tools/README.md`](tools/README.md)。
+
+### 原始数据（`data/`）
+
+| 文件 | 说明 |
+|------|------|
+| [`data/search_results.json`](data/search_results.json) | arXiv + GitHub 历史检索缓存（ZED2i 相关，含 2026-04/05 最新论文） |
+| [`data/crowdroom_papers_raw.json`](data/crowdroom_papers_raw.json) | CrowdRoom 专属抓取结果（当前因 429 限速为空） |
+| [`data/lyra2_paper.txt`](data/lyra2_paper.txt) | Lyra 2.0 论文全文（arXiv:2604.13036） |
+| [`data/lyra2_paper.pdf`](data/lyra2_paper.pdf) | Lyra 2.0 论文 PDF |
+
+详见 [`data/README.md`](data/README.md)。
+
+---
+
 ## 文档统计
 
 | 主题分类 | 文档数 | 覆盖论文数（估计） |
@@ -101,7 +127,7 @@ description: "研究论文调研综述目录，涵盖世界模型、空间感知
 | 物理理解 | 1 | ~20 |
 | 传感器与数据管道 | 2 | ~135 |
 | 项目相关（CrowdRoom） | 1 | 0（API 限流，骨架版） |
-| **合计** | **9** | **~260** |
+| **合计** | **8** | **~260** |
 
 ---
 
diff --git a/research/tools/README.md b/research/tools/README.md
new file mode 100644
index 0000000..daa67a0
--- /dev/null
+++ b/research/tools/README.md
@@ -0,0 +1,39 @@
+# Research Tools
+
+本目录存放论文调研相关的自动化脚本工具。
+
+## 脚本说明
+
+| 脚本 | 用途 |
+|------|------|
+| `search_info.py` | arXiv + GitHub 主题化检索工具，输出 `../data/search_results.json` |
+| `fetch_crowdroom_papers.py` | CrowdRoom 专属 arXiv 论文抓取，输出 `../data/crowdroom_papers_raw.json` |
+| `gen_review_from_json.py` | 将 `search_results.json` 渲染为 ZED2i 综述 Markdown |
+| `_build_crowdroom_review.py` | 将抓取数据构建为 CrowdRoom 综述 Markdown |
+
+## 使用方法
+
+```bash
+# 1. 抓取 ZED2i 相关论文
+cd /path/to/worldmodel
+HTTPS_PROXY=http://127.0.0.1:6984 python3 research/tools/search_info.py \
+    --proxy http://127.0.0.1:6984 \
+    --max-results 15 --delay 5.0 \
+    --out research/data/search_results.json
+
+# 2. 生成 ZED2i 综述
+python3 research/tools/gen_review_from_json.py \
+    --in research/data/search_results.json \
+    --out research/spatial-memory/zed2i_arxiv_live_review.md
+
+# 3. 抓取 CrowdRoom 相关论文
+python3 research/tools/fetch_crowdroom_papers.py \
+    --max-results 20 --delay 8 --max-days 365 \
+    --out research/data/crowdroom_papers_raw.json
+
+# 4. 生成 CrowdRoom 综述
+python3 research/tools/_build_crowdroom_review.py \
+    --fresh research/data/crowdroom_papers_raw.json \
+    --legacy research/data/search_results.json \
+    --out research/crowdroom/crowdroom_related_papers_2026.md
+```
diff --git a/research/_build_crowdroom_review.py b/research/tools/_build_crowdroom_review.py
similarity index 99%
rename from research/_build_crowdroom_review.py
rename to research/tools/_build_crowdroom_review.py
index 381180b..4cd1353 100644
--- a/research/_build_crowdroom_review.py
+++ b/research/tools/_build_crowdroom_review.py
@@ -347,9 +347,9 @@ def render_paper_entry(p, sc, idx):
 
 def main():
     ap = argparse.ArgumentParser()
-    ap.add_argument("--fresh", default="research/crowdroom_papers_raw.json")
-    ap.add_argument("--legacy", default="research/search_results.json")
-    ap.add_argument("--out", default="research/crowdroom_related_papers_2026.md")
+    ap.add_argument("--fresh", default="research/data/crowdroom_papers_raw.json")
+    ap.add_argument("--legacy", default="research/data/search_results.json")
+    ap.add_argument("--out", default="research/crowdroom/crowdroom_related_papers_2026.md")
     ap.add_argument("--threshold", type=float, default=4.0)
     ap.add_argument("--max-per-topic", type=int, default=8)
     ap.add_argument("--include-curated", action="store_true", default=True,
diff --git a/research/fetch_crowdroom_papers.py b/research/tools/fetch_crowdroom_papers.py
similarity index 99%
rename from research/fetch_crowdroom_papers.py
rename to research/tools/fetch_crowdroom_papers.py
index 0967faf..6b2dbae 100644
--- a/research/fetch_crowdroom_papers.py
+++ b/research/tools/fetch_crowdroom_papers.py
@@ -180,7 +180,7 @@ def parse_args() -> argparse.Namespace:
     ap.add_argument("--max-results", type=int, default=20)
     ap.add_argument("--delay", type=float, default=5.0, help="主题间延迟秒数，arXiv ≥3s")
     ap.add_argument("--topics", nargs="+", default=None)
-    ap.add_argument("--out", default="research/crowdroom_papers_raw.json")
+    ap.add_argument("--out", default="research/data/crowdroom_papers_raw.json")
     ap.add_argument("--max-days", type=int, default=365, help="仅保留近 N 天提交的论文；0 不过滤")
     ap.add_argument("--sort", choices=["submittedDate", "relevance"], default="submittedDate")
     return ap.parse_args()
diff --git a/research/gen_review_from_json.py b/research/tools/gen_review_from_json.py
similarity index 98%
rename from research/gen_review_from_json.py
rename to research/tools/gen_review_from_json.py
index 00401c7..ad6c653 100644
--- a/research/gen_review_from_json.py
+++ b/research/tools/gen_review_from_json.py
@@ -147,8 +147,8 @@ def render_github_topic(topic: str, repos: list[dict]) -> list[str]:
 
 def main() -> int:
     ap = argparse.ArgumentParser()
-    ap.add_argument("--in", dest="inp", default="research/search_results.json")
-    ap.add_argument("--out", default="research/zed2i_arxiv_live_review.md")
+    ap.add_argument("--in", dest="inp", default="research/data/search_results.json")
+    ap.add_argument("--out", default="research/spatial-memory/zed2i_arxiv_live_review.md")
     args = ap.parse_args()
 
     with open(args.inp, "r", encoding="utf-8") as f:
diff --git a/research/search_info.py b/research/tools/search_info.py
similarity index 99%
rename from research/search_info.py
rename to research/tools/search_info.py
index 380e0dd..2697d33 100644
--- a/research/search_info.py
+++ b/research/tools/search_info.py
@@ -258,7 +258,7 @@ def main() -> int:
     parser = argparse.ArgumentParser(description="arXiv + GitHub 主题化检索")
     parser.add_argument("--proxy", default=None, help="代理 URL，如 http://127.0.0.1:6984")
     parser.add_argument("--max-results", type=int, default=15, help="每个主题最多论文数")
-    parser.add_argument("--out", default="research/search_results.json", help="输出 JSON 路径")
+    parser.add_argument("--out", default="research/data/search_results.json", help="输出 JSON 路径")
     parser.add_argument(
         "--topics",
         nargs="+",