haive.core.engine.retriever.providers.RePhraseQueryRetrieverConfig

Rephrase Query Retriever implementation for the Haive framework.

from typing import Any This module provides a configuration class for the Rephrase Query retriever, which reformulates user queries using an LLM to improve retrieval performance by creating more effective search queries.

The RePhraseQueryRetriever works by: 1. Taking the user’s original query as input 2. Using an LLM to rephrase the query for better search effectiveness 3. Running the rephrased query against the base retriever 4. Returning documents found using the improved query

This retriever is particularly useful when: - User queries are poorly formulated or ambiguous - Need to improve search effectiveness through query optimization - Building systems that need to handle natural language queries better - Want to bridge the gap between user intent and retrieval effectiveness

The implementation integrates with LangChain’s RePhraseQueryRetriever while providing a consistent Haive configuration interface with LLM integration.

Classes

RePhraseQueryRetrieverConfig

Configuration for Rephrase Query retriever in the Haive framework.

Module Contents

class haive.core.engine.retriever.providers.RePhraseQueryRetrieverConfig.RePhraseQueryRetrieverConfig

Bases: haive.core.engine.retriever.retriever.BaseRetrieverConfig

Configuration for Rephrase Query retriever in the Haive framework.

This retriever reformulates user queries using an LLM to improve retrieval performance by creating more effective search queries.

retriever_type

The type of retriever (always REPHRASE_QUERY).

Type:

RetrieverType

base_retriever

The underlying retriever to query with rephrased query.

Type:

BaseRetrieverConfig

llm_config

LLM configuration for query rephrasing.

Type:

AugLLMConfig

prompt_template

Custom prompt template for rephrasing.

Type:

Optional[str]

Examples

>>> from haive.core.engine.retriever import RePhraseQueryRetrieverConfig
>>> from haive.core.engine.retriever.providers.VectorStoreRetrieverConfig import VectorStoreRetrieverConfig
>>> from haive.core.engine.aug_llm import AugLLMConfig
>>>
>>> # Create base retriever and LLM config
>>> base_config = VectorStoreRetrieverConfig(name="base", vectorstore_config=vs_config)
>>> llm_config = AugLLMConfig(model_name="gpt-3.5-turbo", provider="openai")
>>>
>>> # Create rephrase query retriever
>>> config = RePhraseQueryRetrieverConfig(
...     name="rephrase_retriever",
...     base_retriever=base_config,
...     llm_config=llm_config
... )
>>>
>>> # Instantiate and use the retriever
>>> retriever = config.instantiate()
>>> docs = retriever.get_relevant_documents("machine learning stuff")

get_input_fields()

Return input field definitions for Rephrase Query retriever.

Return type:

dict[str, tuple[type, Any]]

get_output_fields()

Return output field definitions for Rephrase Query retriever.

Return type:

dict[str, tuple[type, Any]]

instantiate()

Create a Rephrase Query retriever from this configuration.

Returns:

Instantiated retriever ready for query rephrasing retrieval.

Return type:

RePhraseQueryRetriever

Raises:

• ImportError – If required packages are not available.

• ValueError – If configuration is invalid.