Package com.tnsai.llm.cache

Enum Class EmbeddingTask

java.lang.Object

java.lang.Enum<EmbeddingTask>

com.tnsai.llm.cache.EmbeddingTask

All Implemented Interfaces:

Serializable, Comparable<EmbeddingTask>, Constable

public enum EmbeddingTask
extends Enum<EmbeddingTask>

The task an embedding is generated for.

Instruction-tuned embedding models (e.g. nomic-embed-text) are contrastively trained with a per-task instruction prefix and condition the vector on it — the same text embedded for two different tasks yields two different vectors. Picking the right task is what pulls related text apart from unrelated text; embedding such a model with no task at all collapses every vector into one degenerate sub-space.

Symmetric vs asymmetric

• CLUSTERING and CLASSIFICATION are symmetric: like is compared with like (prompt↔prompt similarity, dedup, semantic textual similarity, the semantic cache). Both sides of a comparison must use the same task or the vectors are not comparable.
• QUERY and DOCUMENT are the asymmetric retrieval pair: a short query is embedded as QUERY and matched against corpus text embedded as DOCUMENT (the RAG / vector-store path).

Providers whose model has no task convention (e.g. HashEmbeddingProvider, OpenAI text-embedding-3) ignore this and return the same vector regardless — see the task-free defaults on EmbeddingProvider.

Since:

0.13.0

See Also:

• EmbeddingProvider.embed(String, EmbeddingTask)

Nested Class Summary

Nested classes/interfaces inherited from class java.lang.Enum

Enum.EnumDesc<E extends Enum<E>>

Enum Constant Summary

Enum Constants

Enum Constant	Description
CLASSIFICATION	Symmetric classification / label assignment.
CLUSTERING	Symmetric similarity: clustering, deduplication, semantic textual similarity, semantic cache.
DOCUMENT	Asymmetric retrieval: corpus text to be retrieved by a QUERY.
QUERY	Asymmetric retrieval: a search query, matched against DOCUMENT vectors.

Method Summary

Modifier and Type	Method	Description
static EmbeddingTask	valueOf(String name)	Returns the enum constant of this class with the specified name.
static EmbeddingTask[]	values()	Returns an array containing the constants of this enum class, in the order they are declared.

Methods inherited from class java.lang.Enum

compareTo, describeConstable, equals, getDeclaringClass, hashCode, name, ordinal, toString, valueOf

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Enum Constant Details

QUERY

public static final EmbeddingTask QUERY

Asymmetric retrieval: a search query, matched against DOCUMENT vectors.

DOCUMENT

public static final EmbeddingTask DOCUMENT

Asymmetric retrieval: corpus text to be retrieved by a QUERY. Also the neutral default for embedding a single standalone piece of text.

CLUSTERING

public static final EmbeddingTask CLUSTERING

Symmetric similarity: clustering, deduplication, semantic textual similarity, semantic cache.

CLASSIFICATION

public static final EmbeddingTask CLASSIFICATION

Symmetric classification / label assignment.

Method Details

values

public static EmbeddingTask[] values()

Returns an array containing the constants of this enum class, in the order they are declared.

Returns:

an array containing the constants of this enum class, in the order they are declared

valueOf

public static EmbeddingTask valueOf(String name)

Returns the enum constant of this class with the specified name. The string must match exactly an identifier used to declare an enum constant in this class. (Extraneous whitespace characters are not permitted.)

Parameters:

name - the name of the enum constant to be returned.

Returns:

the enum constant with the specified name

Throws:

IllegalArgumentException - if this enum class has no constant with the specified name

NullPointerException - if the argument is null