diff --git a/docs/en/notes/mm_guide/image_understanding/image_gcot.md b/docs/en/notes/mm_guide/image_understanding/image_gcot.md
index 85a41201..96be0e2a 100644
--- a/docs/en/notes/mm_guide/image_understanding/image_gcot.md
+++ b/docs/en/notes/mm_guide/image_understanding/image_gcot.md
@@ -39,13 +39,10 @@ cd run_gcot
 
 Save the code in the "Pipeline Example" section below as `image_gcot_pipeline.py`.
 
-### Step 3: Configure Parameters
-
-Ensure you have a VLM model capable of grounding (e.g., Qwen2.5-VL-7B-Instruct).
+### Step 3: Download Example Data
 
 ```bash
-# Install dependencies
-pip install open-dataflow vllm
+huggingface-cli download --repo-type dataset OpenDCAI/dataflow-demo-image --local-dir example_data
 
 ```
 
diff --git a/docs/en/notes/mm_guide/image_understanding/image_scale_caption_pipeline.md b/docs/en/notes/mm_guide/image_understanding/image_scale_caption_pipeline.md
index 8dc770ae..aeff0c1d 100644
--- a/docs/en/notes/mm_guide/image_understanding/image_scale_caption_pipeline.md
+++ b/docs/en/notes/mm_guide/image_understanding/image_scale_caption_pipeline.md
@@ -40,13 +40,10 @@ cd run_scalecap
 
 Save the code in the "Pipeline Example" section below as `scalecap_pipeline.py`.
 
-### Step 3: Configure Parameters
-
-Ensure the VLM model path (e.g., Qwen2.5-VL) is correct.
+### Step 3: Download Example Data
 
 ```bash
-# Install dependencies
-pip install open-dataflow vllm
+huggingface-cli download --repo-type dataset OpenDCAI/dataflow-demo-image --local-dir example_data
 
 ```
 
diff --git a/docs/en/notes/mm_guide/image_understanding/image_visual_only_mcq_pipeline.md b/docs/en/notes/mm_guide/image_understanding/image_visual_only_mcq_pipeline.md
index cc3806af..f6c04ad8 100644
--- a/docs/en/notes/mm_guide/image_understanding/image_visual_only_mcq_pipeline.md
+++ b/docs/en/notes/mm_guide/image_understanding/image_visual_only_mcq_pipeline.md
@@ -36,13 +36,10 @@ cd run_vis_mcq
 
 Save the code in the "Pipeline Example" section below as `visual_mcq_pipeline.py`.
 
-### Step 3: Configure Parameters
-
-Control filtering thresholds via CLI. For example, requiring 100% visual accuracy and less than 25% blind accuracy:
+### Step 3: Download Example Data
 
 ```bash
-# Install dependencies
-pip install open-dataflow vllm
+huggingface-cli download --repo-type dataset OpenDCAI/dataflow-demo-image --local-dir example_data
 
 ```
 
diff --git a/docs/en/notes/mm_guide/image_understanding/vision_mct_reasoning_pipeline.md b/docs/en/notes/mm_guide/image_understanding/vision_mct_reasoning_pipeline.md
index 0904691e..c559087b 100644
--- a/docs/en/notes/mm_guide/image_understanding/vision_mct_reasoning_pipeline.md
+++ b/docs/en/notes/mm_guide/image_understanding/vision_mct_reasoning_pipeline.md
@@ -39,13 +39,10 @@ cd run_mcts_reasoning
 
 Save the code in the "Pipeline Example" section below as `vision_mcts_pipeline.py`.
 
-### Step 3: Configure Parameters
-
-Ensure the input file (jsonl) contains the `tree` field (for extraction) or just `question/image` (for generation).
+### Step 3: Download Example Data
 
 ```bash
-# Install dependencies
-pip install open-dataflow vllm
+huggingface-cli download --repo-type dataset OpenDCAI/dataflow-demo-image --local-dir example_data
 
 ```