Add new experimental rank_vectors mapping for late-interaction second order ranking #118804
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Documentation preview: |
|
Pinging @elastic/es-search-relevance (Team:Search Relevance) |
|
Hi @benwtrent, I've created a changelog YAML for you. Note that since this PR is labelled |
benwtrent
commented
Dec 16, 2024
.../src/main/java/org/elasticsearch/xpack/rank/vectors/script/RankVectorsPainlessExtension.java
Show resolved
Hide resolved
carlosdelest
approved these changes
Dec 18, 2024
| private final DenseVectorFieldMapper.ElementType elementType; | ||
| private final int dims; | ||
|
|
||
| RankVectorsDVLeafFieldData(LeafReader reader, String field, DenseVectorFieldMapper.ElementType elementType, int dims) { |
There was a problem hiding this comment.
We really should make ElementType its own class. We should start some refactoring of DenseVectorFieldMapper after this PR lands.
…asticsearch into feature/rank-vectors-plugin
💚 Backport successful
|
benwtrent
added a commit
to benwtrent/elasticsearch
that referenced
this pull request
Jan 6, 2025
… order ranking (elastic#118804) Late-interaction models are powerful rerankers. While their size and overall cost doesn't lend itself for HNSW indexing, utilizing them as second order "brute-force" reranking can provide excellent boosts in relevance. At generally lower inference times than large cross-encoders. This commit exposes a new experimental `rank_vectors` field that allows for maxSim operations. This unlocks the initial, and most common use of late-interaction dense-models. For example, this is how you would use it via the API: ``` PUT index { "mappings": { "properties": { "late_interaction_vectors": { "type": "rank_vectors" } } } } ``` Then to index: ``` POST index/_doc { "late_interaction_vectors": [[0.1, ...],...] } ``` For querying, scoring can be exposed with scripting: ``` POST index/_search { "query": { "script_score": { "query": { "match_all": {} }, "script": { "source": "maxSimDotProduct(params.query_vector, 'my_vector')", "params": { "query_vector": [[0.42, ...], ...] } } } } } ``` Of course, the initial ranking should be done before re-scoring or combining via the `rescore` parameter, or simply passing whatever first phase retrieval you want as the inner query in `script_score`.
elasticsearchmachine
pushed a commit
that referenced
this pull request
Jan 6, 2025
…second order ranking (#118804) (#119601) * Add new experimental rank_vectors mapping for late-interaction second order ranking (#118804) Late-interaction models are powerful rerankers. While their size and overall cost doesn't lend itself for HNSW indexing, utilizing them as second order "brute-force" reranking can provide excellent boosts in relevance. At generally lower inference times than large cross-encoders. This commit exposes a new experimental `rank_vectors` field that allows for maxSim operations. This unlocks the initial, and most common use of late-interaction dense-models. For example, this is how you would use it via the API: ``` PUT index { "mappings": { "properties": { "late_interaction_vectors": { "type": "rank_vectors" } } } } ``` Then to index: ``` POST index/_doc { "late_interaction_vectors": [[0.1, ...],...] } ``` For querying, scoring can be exposed with scripting: ``` POST index/_search { "query": { "script_score": { "query": { "match_all": {} }, "script": { "source": "maxSimDotProduct(params.query_vector, 'my_vector')", "params": { "query_vector": [[0.42, ...], ...] } } } } } ``` Of course, the initial ranking should be done before re-scoring or combining via the `rescore` parameter, or simply passing whatever first phase retrieval you want as the inner query in `script_score`. * Update docs/changelog/119601.yaml
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Late-interaction models are powerful rerankers. While their size and overall cost doesn't lend itself for HNSW indexing, utilizing them as second order "brute-force" reranking can provide excellent boosts in relevance. At generally lower inference times than large cross-encoders.
This commit exposes a new experimental
rank_vectorsfield that allows for maxSim operations. This unlocks the initial, and most common use of late-interaction dense-models.For example, this is how you would use it via the API:
Then to index:
For querying, scoring can be exposed with scripting:
Of course, the initial ranking should be done before re-scoring or combining via the
rescoreparameter, or simply passing whatever first phase retrieval you want as the inner query inscript_score.