[
https://issues.apache.org/jira/browse/SOLR-17525?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17907445#comment-17907445
]
ASF subversion and git services commented on SOLR-17525:
--------------------------------------------------------
Commit 0abf40d2ccefbb551335240a28efdd1dc5d1d214 in solr's branch
refs/heads/branch_9x from Alessandro Benedetti
[ https://gitbox.apache.org/repos/asf?p=solr.git;h=0abf40d2cce ]
SOLR-17525 backport to 9.x (#2921)
* SOLR-17525: Text Embedder Query Parser (#2809)
* added support to store Large Language Models external endpoints config for
text to vector encoding
* added a query parser that can use such external models to encode text to
vector and run a KNearest Neighbour search
---------
Co-authored-by: Christine Poerschke <[email protected]>
Co-authored-by: Christos Malliaridis <[email protected]>
Co-authored-by: David Smiley <[email protected]>
(cherry picked from commit e8803e6f707b4600cec1e7e70f1d52f895bae0a6)
> Text to Vector Query Parser
> ---------------------------
>
> Key: SOLR-17525
> URL: https://issues.apache.org/jira/browse/SOLR-17525
> Project: Solr
> Issue Type: New Feature
> Security Level: Public(Default Security Level. Issues are Public)
> Components: query parsers
> Reporter: Alessandro Benedetti
> Priority: Major
> Labels: pull-request-available
> Time Spent: 21h 40m
> Remaining Estimate: 0h
>
> Scope of this issue is to introduce support for automatic text vectorisation
> in Apache Solr, directly in a query parser.
> A LLM fine-tuned for sentence similarity will be accessed to embed the text.
> Apache Solr will host the configuration parameters to access embedding
> services and the query parser will use such services to directly encode the
> query to a vector and then run vector search on a DenseVectorField.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
