diff --git a/.gitignore b/.gitignore
index 0a23ae0..a24c778 100644
--- a/.gitignore
+++ b/.gitignore
@@ -37,6 +37,7 @@ examples/hBN/negf_output_k70/self_energy/*
 examples/CNT/output/*
 examples/long_cnt/*
 CLAUDE*
+ai_docs/*
 
 
 run_config.json
diff --git a/Dockerfile b/Dockerfile
index 7de2431..1e020c2 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -31,22 +31,22 @@ RUN apt-get update > /dev/null && \
 WORKDIR /app
 COPY . .
 
-# 2. 创建环境并安装所有依赖
+# 2. Create the environment and install all dependencies
 RUN \
     sed -i 's/build-backend = "poetry_dynamic_versioning.backend"/build-backend = "poetry.core.masonry.api"/' pyproject.toml && \
     conda create -n dpnegf python=3.10 -c conda-forge -y && \
     git clone https://github.com/deepmodeling/DeePTB.git && \
     conda run -n dpnegf pip install --upgrade pip setuptools wheel && \
-    # [1] 强制拉取纯 CPU 版本的 PyTorch 2.1.1，极大地减小镜像体积并对齐底层接口
-    conda run -n dpnegf pip install torch==2.1.1 --index-url https://download.pytorch.org/whl/cpu && \
-    # [2] 强制使用 PyG 专属源拉取 torch-scatter，并使用 --only-binary=torch-scatter 彻底关闭源码编译。
-    # 这样如果找不到精确匹配的 Wheel，它会立刻报错，而不是花 10 分钟编译出一个会引发崩溃的包。
-    conda run -n dpnegf pip install torch-scatter -f https://data.pyg.org/whl/torch-2.1.1+cpu.html --only-binary=torch-scatter && \
-    # [3] 给本地仓库的安装加上 CPU 源保护，防止安装过程触发隐藏依赖，把刚才装好的 CPU 版 Torch 顶替成带 CUDA 的版本
-    conda run -n dpnegf pip install ./DeePTB torch==2.1.1 --extra-index-url https://download.pytorch.org/whl/cpu && \
-    conda run -n dpnegf pip install ./ torch==2.1.1 --extra-index-url https://download.pytorch.org/whl/cpu && \
+    # [1] Force the CPU-only PyTorch 2.5.1 wheel to shrink the image and align the low-level ABI.
+    conda run -n dpnegf pip install torch==2.5.1 --index-url https://download.pytorch.org/whl/cpu && \
+    # [2] Pull torch-scatter from the PyG wheel index and use --only-binary=torch-scatter to fully disable source builds.
+    # If no matching wheel is found it fails immediately instead of spending ~10 minutes compiling a package that would crash at runtime.
+    conda run -n dpnegf pip install torch-scatter -f https://data.pyg.org/whl/torch-2.5.0+cpu.html --only-binary=torch-scatter && \
+    # [3] Guard the local-repo installs with the CPU index so hidden dependencies can't replace the CPU torch with a CUDA build.
+    conda run -n dpnegf pip install ./DeePTB torch==2.5.1 --extra-index-url https://download.pytorch.org/whl/cpu && \
+    conda run -n dpnegf pip install ./ torch==2.5.1 --extra-index-url https://download.pytorch.org/whl/cpu && \
     conda clean --all -y && \
     rm -rf /root/.cache/pip
 
-# 3. 设置默认启动环境
+# 3. Activate the dpnegf environment by default in interactive shells
 RUN echo "conda activate dpnegf" >> ~/.bashrc