langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-23 04:55:28 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

multiple conversation, how to seach vector with all the user input #810

New Issue
Closed
opened 2026-02-21 17:28:32 -05:00 by yindo · 5 comments
yindo commented 2026-02-21 17:28:32 -05:00
Owner
Copy Link
Copy Source

Originally created by @hunter-eric on GitHub (Dec 13, 2023).

Originally assigned to: @JohnJyong on GitHub.

Self Checks

Dify version

0.3.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

first convesation i search "BYD company" , second conversation i search "what was it's advantages",the response of vector search was not about "BYD"

✔️ Expected Behavior

all response of vector search was about all the user input

❌ Actual Behavior

the response of vector search was about user intends

Originally created by @hunter-eric on GitHub (Dec 13, 2023). Originally assigned to: @JohnJyong on GitHub. ### Self Checks - [X] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [X] I confirm that I am using English to file this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). ### Dify version 0.3.2 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce first convesation i search "BYD company" , second conversation i search "what was it's advantages",the response of vector search was not about "BYD" ### ✔️ Expected Behavior all response of vector search was about all the user input ### ❌ Actual Behavior the response of vector search was about user intends
yindo added the 💪 enhancement label 2026-02-21 17:28:32 -05:00
yindo closed this issue 2026-02-21 17:28:32 -05:00
yindo commented 2026-02-21 17:28:32 -05:00
Author
Owner
Copy Link
Copy Source

@guchenhe commented on GitHub (Dec 14, 2023):

Retrieval is done on a per-query basis by design, so it's expected that your second search won't return documents relevant to your first search.

However, in an actual conversation the model does have knowledge on your previous queries and retrieval results, and will respond accordingly (how well it retains previous context also depends on how you've written your prompts & context window)

@guchenhe commented on GitHub (Dec 14, 2023): Retrieval is done on a per-query basis by design, so it's expected that your second search won't return documents relevant to your first search. However, in an actual conversation the model does have knowledge on your previous queries and retrieval results, and will respond accordingly (how well it retains previous context also depends on how you've written your prompts & context window)
yindo commented 2026-02-21 17:28:32 -05:00
Author
Owner
Copy Link
Copy Source

@hunter-eric commented on GitHub (Dec 15, 2023):

Retrieval is done on a per-query basis by design, so it's expected that your second search won't return documents relevant to your first search.

However, in an actual conversation the model does have knowledge on your previous queries and retrieval results, and will respond accordingly (how well it retains previous context also depends on how you've written your prompts & context window)

we found if we use gtp4 in dify, it will change the second search query automaticlly, that very cool. But it won't work when use other llm model.

@hunter-eric commented on GitHub (Dec 15, 2023): > Retrieval is done on a per-query basis by design, so it's expected that your second search won't return documents relevant to your first search. > > However, in an actual conversation the model does have knowledge on your previous queries and retrieval results, and will respond accordingly (how well it retains previous context also depends on how you've written your prompts & context window) we found if we use gtp4 in dify, it will change the second search query automaticlly, that very cool. But it won't work when use other llm model.
yindo commented 2026-02-21 17:28:32 -05:00
Author
Owner
Copy Link
Copy Source

@VincePotato commented on GitHub (Jan 16, 2024):

We use chat history as input context for chat model to better understand user query. That is not achieved through rewriting the user query.

@VincePotato commented on GitHub (Jan 16, 2024): We use chat history as input context for chat model to better understand user query. That is not achieved through rewriting the user query.
yindo commented 2026-02-21 17:28:32 -05:00
Author
Owner
Copy Link
Copy Source

@hunter-eric commented on GitHub (Jan 16, 2024):

We use chat history as input context for chat model to better understand user query. That is not achieved through rewriting the user query.

this could not take advantage of Vector search，because chat history is not similar to vector data

@hunter-eric commented on GitHub (Jan 16, 2024): > We use chat history as input context for chat model to better understand user query. That is not achieved through rewriting the user query. this could not take advantage of Vector search，because chat history is not similar to vector data
yindo commented 2026-02-21 17:28:32 -05:00
Author
Owner
Copy Link
Copy Source

@github-actions[bot] commented on GitHub (Jan 31, 2024):

Close due to it's no longer active, if you have any questions, you can reopen it.

@github-actions[bot] commented on GitHub (Jan 31, 2024): Close due to it's no longer active, if you have any questions, you can reopen it.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Knowledge Pipeline
Proposal
WIP
cloud
dependencies
feat:webapp
good first issue
outdated
performance
pull-request
Mirrored from GitHub Pull Request
 python:uv
refactor
size:L
stale
status: accepting prs
status: blocked
status: in discussion
stt & tts
vulnerablity
web
wont-fix
🌊 feat:workflow
🌚 invalid
🐞 bug
👻 feat:rag
💪 enhancement
📚 documentation
🔨 feat:tools
🙋‍♂️ question
🙏 help wanted
🤔 cant-reproduce
🤖 feat:agent
No Label 💪 enhancement
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#810
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?