Some states are lost after using LangGraph with human-in-the-loop, especially when there is a lot of content in the state.

[Smile0524]

Checked other resources

• This is a bug, not a usage question. For questions, please use the LangChain Forum (https://forum.langchain.com/).
• I added a clear and detailed title that summarizes the issue.
• I read what a minimal reproducible example is (https://stackoverflow.com/help/minimal-reproducible-example).
• I included a self-contained, minimal example that demonstrates the issue INCLUDING all the relevant imports. The code run AS IS to reproduce the issue.

Example Code

use AsyncPostgresSaver
use interrupt

Error Message and Stack Trace (if applicable)

When there is less content in "messages", the complete "messages" of state from the previous round can be obtained after the first human-in-loop. When there is more content in "messages", the "messages" of state from the previous round is empty.

Description

When there is less content in "messages", the complete "messages" of state from the previous round can be obtained after the first human-in-loop. When there is more content in "messages", the "messages" of state from the previous round is empty.
After 1st human in the loop, the same thread_id can not cannot obtain the previous complete state.

System Info

When there is less content in "messages", the complete "messages" of state from the previous round can be obtained after the first human-in-loop. When there is more content in "messages", the "messages" of state from the previous round is empty.
After 1st human in the loop, the same thread_id can not cannot obtain the previous complete state.