Polish docs

docs/tutorials/chat-on-tabular-data.md

@@ -1,6 +1,13 @@
 # Chat on Tabular Data
 
-Beyond simple conversations, `tablegpt-agent` can also analyze and process tablular data. For data analysis, preprocess the data first with TableGPT. Use a fixed session_id to keep the agent within the same execution context, and set a save point to enable memory retention:
+TableGPT Agent excels at analyzing and processing tabular data. To perform data analysis, you need to first let the agent "see" the dataset. This is done by a specific "file-reading" workflow. In short, you begin by "uploading" the dataset and let the agent read it. Once the data is read, you can ask the agent questions about it.
+
+> To learn more about the file-reading workflow, see [File Reading](./explanation/file-reading.md).
+
+For data analysis tasks, we introduce two important parameters when creating the agent: `checkpointer` and `session_id`.
+
+- The `checkpointer` should be an instance of `langgraph.checkpoint.base.BaseCheckpointSaver`, which acts as a versioned "memory" for the agent. (See [langgraph's persistence concept](https://langchain-ai.github.io/langgraph/concepts/persistence) for more details.)
+- The `session_id` is a unique identifier for the current session. It ties the agent's execution to a specific kernel, ensuring that the agent's results are retained across multiple invocations.
 
 ```pycon
 >>> from langgraph.checkpoint.memory import MemorySaver

docs/tutorials/quickstart.md

@@ -2,7 +2,7 @@
 
 ## Installation
 
-To install `tablegpt-agent`, use the following command:
+To install TableGPT Agent, use the following command:
 
 ```sh
 pip install tablegpt-agent
@@ -16,14 +16,14 @@ pip install tablegpt-agent[local]
 
 ## Setup LLM Service
 
-Before using `tablegpt-agent`, ensure that you have an OpenAI-compatible server set up to host TableGPT2. We recommend using [vllm](https://github.com/vllm-project/vllm) for this:
+Before using TableGPT Agent, ensure that you have an OpenAI-compatible server set up to host TableGPT2. We recommend using [vllm](https://github.com/vllm-project/vllm) for this:
 
 > **Note:**  If you need to use `tablegpt-agent` to analyze tabular data, please ensure your `vllm>=0.5.5`
+
 ```sh
 pip install 'vllm>=0.5.5'
 ```
 
-
 ```sh
 python -m vllm.entrypoints.openai.api_server --served-model-name TableGPT2-7B --model path/to/weights
 ```
@@ -32,8 +32,9 @@ python -m vllm.entrypoints.openai.api_server --served-model-name TableGPT2-7B --
 
 ## Chat with TableGPT Agent
 
-To create a `TablegptAgent`, you'll need both an `LLM` and a `PyBoxManager` instance:
+To create an agent, you'll need both an `LLM` and a `PyBoxManager` instance:
 > **NOTE** The `llm` is created using `langchain-openai`, please install it first.
+
 ```sh
 pip install langchain-openai
 ```