From 1257c30e1a56141da93a43179bdb130e8e24cc3e Mon Sep 17 00:00:00 2001 From: dengjiaji Date: Mon, 15 Sep 2025 15:44:23 +0800 Subject: [PATCH] docs: update experiment tutorials and documentation structure --- README.md | 14 +++++++------- README_ZH.md | 14 +++++++------- {cookbook => docs/cookbook}/appworld/quickstart.md | 0 {cookbook => docs/cookbook}/bfcl/quickstart.md | 0 .../cookbook}/frozenlake/quickstart.md | 0 docs/index.md | 2 +- mkdocs.yml | 8 +++++++- 7 files changed, 22 insertions(+), 16 deletions(-) rename {cookbook => docs/cookbook}/appworld/quickstart.md (100%) rename {cookbook => docs/cookbook}/bfcl/quickstart.md (100%) rename {cookbook => docs/cookbook}/frozenlake/quickstart.md (100%) diff --git a/README.md b/README.md index ff014ea9..d970dfcc 100644 --- a/README.md +++ b/README.md @@ -37,8 +37,8 @@ Personal memory helps "**understand user preferences**", while task memory helps in [MemoryScope](https://github.com/modelscope/Reme/tree/memoryscope_branch). - **[2025-09]** 🧪 We validated the effectiveness of task memory extraction and reuse in agents in appworld, bfcl(v3), and frozenlake environments. For more information, - check [appworld exp](./cookbook/appworld/quickstart.md), [bfcl exp](./cookbook/bfcl/quickstart.md), - and [frozenlake exp](./cookbook/frozenlake/quickstart.md). + check [appworld exp](docs/cookbook/appworld/quickstart.md), [bfcl exp](docs/cookbook/bfcl/quickstart.md), + and [frozenlake exp](docs/cookbook/frozenlake/quickstart.md). - **[2025-08]** 🚀 MCP protocol support is now available -> [MCP Quick Start](docs/mcp_quick_start.md). - **[2025-06]** 🚀 Multiple backend vector storage support (Elasticsearch & ChromaDB) -> [Vector DB quick start](docs/vector_store_api_guide.md). @@ -349,7 +349,7 @@ response = requests.post("http://localhost:8002/retrieve_task_memory", json={ ## 🧪 Experiments -### 🌍 [Appworld Experiment](./cookbook/appworld/quickstart.md) +### 🌍 [Appworld Experiment](docs/cookbook/appworld/quickstart.md) We tested ReMe on Appworld using qwen3-8b: @@ -362,9 +362,9 @@ Pass@K measures the probability that at least one of the K generated samples suc score=1). The current experiment uses an internal AppWorld environment, which may have slight differences. -You can find more details on reproducing the experiment in [quickstart.md](cookbook/appworld/quickstart.md). +You can find more details on reproducing the experiment in [quickstart.md](docs/cookbook/appworld/quickstart.md). -### 🧊 [Frozenlake Experiment](./cookbook/frozenlake/quickstart.md) +### 🧊 [Frozenlake Experiment](docs/cookbook/frozenlake/quickstart.md) | without ReMe | with ReMe | |:--------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------:| @@ -377,9 +377,9 @@ We tested on 100 random frozenlake maps using qwen3-8b: | without ReMe | 0.66 | | with ReMe | 0.72 **(+6.0%)** | -You can find more details on reproducing the experiment in [quickstart.md](cookbook/frozenlake/quickstart.md). +You can find more details on reproducing the experiment in [quickstart.md](docs/cookbook/frozenlake/quickstart.md). -### 🔧 [BFCL-V3 Experiment](./cookbook/bfcl/quickstart.md) +### 🔧 [BFCL-V3 Experiment](docs/cookbook/bfcl/quickstart.md) We tested ReMe on BFCL-V3 multi-turn-base (randomly split 50train/150val) using qwen3-8b: diff --git a/README_ZH.md b/README_ZH.md index dab40858..d1d1365d 100644 --- a/README_ZH.md +++ b/README_ZH.md @@ -33,8 +33,8 @@ ReMe为AI智能体提供了统一的记忆与经验系统——在跨用户、 正式发布,整合任务记忆与个人记忆。如果想使用原始的memoryscope项目,你可以在[MemoryScope](https://github.com/modelscope/Reme/tree/memoryscope_branch) 中找到。 - **[2025-09]** 🧪 我们在appworld, bfcl(v3) - 以及frozenlake环境验证了任务记忆抽取与复用在Agent中的效果,更多信息请查看 [appworld exp](./cookbook/appworld/quickstart.md), [bfcl exp](./cookbook/bfcl/quickstart.md) - 和 [frozenlake exp](./cookbook/frozenlake/quickstart.md)。 + 以及frozenlake环境验证了任务记忆抽取与复用在Agent中的效果,更多信息请查看 [appworld exp](docs/cookbook/appworld/quickstart.md), [bfcl exp](docs/cookbook/bfcl/quickstart.md) + 和 [frozenlake exp](docs/cookbook/frozenlake/quickstart.md)。 - **[2025-08]** 🚀 MCP协议支持已上线-> [MCP指南](docs/mcp_quick_start.md)。 - **[2025-06]** 🚀 多后端向量存储支持 (Elasticsearch & ChromaDB) -> [向量数据库指南](docs/vector_store_api_guide.md)。 - **[2024-09]** 🧠 [MemoryScope](https://github.com/modelscope/Reme/tree/memoryscope_branch) v0.1 发布,个性化和时间感知的记忆存储与使用。 @@ -327,7 +327,7 @@ response = requests.post("http://localhost:8002/retrieve_task_memory", json={ ## 🧪 实验 -### 🌍 [Appworld 实验](cookbook/appworld/quickstart.md) +### 🌍 [Appworld 实验](docs/cookbook/appworld/quickstart.md) 我们在 Appworld 上使用 qwen3-8b 测试 ReMe: @@ -339,10 +339,10 @@ response = requests.post("http://localhost:8002/retrieve_task_memory", json={ Pass@K 衡量的是在生成的 K 个样本中,至少有一个成功完成任务(score=1)的概率。 当前实验使用的是一个内部的 AppWorld 环境,可能存在轻微差异。 -你可以在 [quickstart.md](cookbook/appworld/quickstart.md) 中找到复现实验的更多细节。 +你可以在 [quickstart.md](docs/cookbook/appworld/quickstart.md) 中找到复现实验的更多细节。 -### 🧊 [Frozenlake 实验](cookbook/frozenlake/quickstart.md) +### 🧊 [Frozenlake 实验](docs/cookbook/frozenlake/quickstart.md) | 不使用ReMe | 使用ReMe | |:--------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------:| @@ -355,9 +355,9 @@ Pass@K 衡量的是在生成的 K 个样本中,至少有一个成功完成任 | without ReMe | 0.66 | | with ReMe | 0.72 **(+6.0%)** | -你可以在 [quickstart.md](cookbook/frozenlake/quickstart.md) 中找到复现实验的更多细节。 +你可以在 [quickstart.md](docs/cookbook/frozenlake/quickstart.md) 中找到复现实验的更多细节。 -### 🔧 [BFCL-V3 实验](cookbook/bfcl/quickstart.md) +### 🔧 [BFCL-V3 实验](docs/cookbook/bfcl/quickstart.md) 我们在 BFCL-V3 multi-turn-base (随机划分50train/150val) 上使用 qwen3-8b 测试 ReMe: diff --git a/cookbook/appworld/quickstart.md b/docs/cookbook/appworld/quickstart.md similarity index 100% rename from cookbook/appworld/quickstart.md rename to docs/cookbook/appworld/quickstart.md diff --git a/cookbook/bfcl/quickstart.md b/docs/cookbook/bfcl/quickstart.md similarity index 100% rename from cookbook/bfcl/quickstart.md rename to docs/cookbook/bfcl/quickstart.md diff --git a/cookbook/frozenlake/quickstart.md b/docs/cookbook/frozenlake/quickstart.md similarity index 100% rename from cookbook/frozenlake/quickstart.md rename to docs/cookbook/frozenlake/quickstart.md diff --git a/docs/index.md b/docs/index.md index bf92a359..7871fa5c 100644 --- a/docs/index.md +++ b/docs/index.md @@ -354,7 +354,7 @@ response = requests.post("http://localhost:8002/retrieve_task_memory", json={ ## 🧪 Experiments -### 🌍 [Appworld Experiment](./cookbook/appworld/quickstart.md) +### 🌍 [Appworld Experiment](cookbook/appworld/quickstart.md) We tested ReMe on Appworld using qwen3-8b: diff --git a/mkdocs.yml b/mkdocs.yml index 80058898..d491be5d 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -27,9 +27,15 @@ nav: - Making SOP Memories: sop_memory/making_sop_memories.md - Extensions: - - Mcp: mcp_quick_start.md + - MCP: mcp_quick_start.md - Vector Store: vector_store_api_guide.md + - Experimental Tutorials: + - AppWorld: cookbook/appworld/quickstart.md + - BFCL: cookbook/bfcl/quickstart.md + - FrozenLake: cookbook/frozenlake/quickstart.md + + - Contributions: contribution.md plugins: